阅读说明：本技术 用于检测数据集的方法和设备 (Method and apparatus for detecting data sets ) 是由 福尔克尔·哈姆斯 于 2021-06-22 设计创作，主要内容包括：一种用于检测数据集的方法和设备,该方法包括以下步骤：接收分布有铺路材料的铺路区域的一部分的第一热分布,该第一热分布包括分配给相应测量点的多个温度值；其中,该部分包括第一子部分；分析第一热分布,以检测该部分内的铺路材料的分离点,从而使得能够确定铺路材料的质量,其中,该分析包括：确定彼此相邻布置并且具有预定范围内的温度值的测量点的第一区,其中,第一区与第一子部分相对应,并且其中,第一区至少部分地被具有预定范围之外的温度值的测量点包围；确定第一区的第一分析参数；并且确定子部分的第一质量参数,以将第一质量参数分配给第一分析参数。(A method and apparatus for detecting a data set, the method comprising the steps of: receiving a first thermal profile of a portion of a paving area having paving material distributed thereon, the first thermal profile including a plurality of temperature values assigned to respective measurement points; wherein the portion comprises a first sub-portion; analyzing the first thermal profile to detect a separation point of the paving material within the portion to enable a determination of a quality of the paving material, wherein the analyzing comprises: determining a first zone of measurement points arranged adjacent to each other and having a temperature value within a predetermined range, wherein the first zone corresponds to the first sub-portion, and wherein the first zone is at least partially surrounded by measurement points having temperature values outside the predetermined range; determining a first analytical parameter for the first region; and determining a first quality parameter of the sub-portion to assign the first quality parameter to the first analysis parameter.)

1. A method for detecting the quality of paving material distributed along a paving area using a paving machine, comprising the steps of:

receiving (1100) a first thermal profile of a portion of the paving area over which the paving material is distributed, the first thermal profile comprising a plurality of temperature values assigned to respective measurement points; wherein the portion comprises a first sub-portion;

analyzing (1200) the first thermal profile to detect a point of separation of the paving material within the portion to enable determination of a quality of the paving material, wherein the analyzing (1200) comprises:

-determining a first zone (1210) of the measurement points arranged adjacent to each other and having a temperature value within a predetermined range, wherein the first zone corresponds to the first sub-portion, and wherein the first zone is at least partially surrounded by measurement points having temperature values outside the predetermined range;

-determining a first analysis parameter (1220) of the first region; and is

-determining a first quality parameter (1230) of a sub-portion to assign to said first analysis parameter, wherein said first analysis parameter enables to draw conclusions about the quality of the road surface;

wherein the analyzing (1200) step is based on learning data comprising at least a learning parameter and at least input data, wherein the input data comprises the first analysis parameter or pattern as the first analysis parameter, and wherein the learning parameter comprises the first quality parameter or a type of error of the paving material at the first sub-portion as the first quality parameter.

2. A method according to claim 1, wherein the method comprises the step of storing the first analysis parameter together with the first quality parameter; and/or wherein the method comprises the step of sending the first analysis parameter together with the first quality parameter to store the first analysis parameter and the first quality parameter on a server.

3. The method according to claim 1, wherein the analyzing (1200) step comprises a self-learning algorithm and/or is based on artificial intelligence.

4. The method of claim 1, wherein the first and second light sources are selected from the group consisting of,

wherein the method comprises a step (1220) of selecting a section of an analysis parameter or one or more possible analysis parameters as a sub-step of determining the first analysis parameter; and/or

Wherein the method comprises the step of selecting the analysis parameter or a segment of one or more possible analysis parameters before performing the step of analyzing (1200); and/or

Wherein the analysis parameter is selected from a group comprising at least one of one or more parameters in the group:

orientation of the first zone with respect to the direction of travel;

an average temperature of the temperature values within the first zone;

a relative temperature of the temperature values within the first zone when compared to the temperature values belonging to the measurement points around the first zone;

the size of the first region;

a temperature deviation within the first zone;

the shape of the pattern of the first region;

a second region where there is a measurement point corresponding to a second sub-portion of the portion;

another region of measurement points corresponding to another sub-portion of the portion;

the distance of the first region to the second region or the first region to another region; and

the relative position of the first to second zone or the first to another zone; or

A combination of at least two or more group elements.

5. The method of claim 1, wherein the method further comprises the step of receiving (1100) parameters of the paving machine and/or a configuration of the paving machine.

6. The method of claim 1, wherein the step of determining the first region of the measurement points comprises a step of pattern determination.

7. The method of claim 1, wherein the first quality parameter is a parameter describing a type of error of the paving material at the first sub-portion; and/or

Wherein the step of determining the first quality parameter comprises the sub-step of receiving (1100) instructions or information from an operator of the paving machine.

8. The method according to claim 1, wherein the step of analyzing (1200) is based on learning data comprising at least a decision parameter defining a type or number of decision nodes, wherein the decision parameter depends on a parameter of the road paver or a configuration of the road paver.

9. The method according to claim 1, wherein the method further comprises the step of receiving (1100) at least one instruction of a quality parameter or a type of error assigned to the paving material at the first sub-portion.

10. The method of claim 1, wherein the receiving (1100) and analyzing (1200) steps are repeated for the same paving area or another paving area, and/or wherein the receiving (1100) and analyzing (1200) steps are repeated for another paving machine.

11. The method of claim 10, wherein repeating enables determining a plurality of comparable data sets for comparable scenarios and/or repeating enables determining a plurality of different data sets for different scenarios.

12. The method of claim 1, wherein the plurality of measurement points within the thermal profile are arranged according to a regular grid.

13. The method according to claim 1, wherein the method is performed for a plurality of temperature profiles or wherein the method is performed for a plurality of temperature profiles overlapping each other.

14. The method of claim 1, wherein the method further comprises the step of outputting instructions to paving machine operators based on the first quality parameter and/or performing actions or changing parameters of a work machine based on the quality parameter; and/or

Wherein the method further comprises the step of determining the first analysis parameter after outputting the instruction or after performing the action or after changing the parameter.

15. The method of claim 1, wherein the first quality parameter characterizes a quality of the road surface.

16. A computer-readable digital storage medium having stored thereon a computer program having a program code for executing the method according to claim 1.

17. An apparatus for detecting the quality of paving material distributed along a paving area, the apparatus comprising:

an interface (1410) for receiving (1100) a first thermal profile of a portion of the paving area over which the paving material is distributed, the first thermal profile comprising a plurality of temperature values assigned to respective measurement points; wherein the portion comprises a first sub-portion; and

a calculation unit (1400) for analyzing (1200) the first thermal profile to detect a separation point of the paving material within the portion to enable determination of a quality of the paving material, wherein the analyzing (1200) comprises:

-determining a first zone of the measurement points arranged adjacent to each other and having a temperature value within a predetermined range, wherein the first zone corresponds to the first sub-portion, and wherein the first zone is at least partially surrounded by measurement points having temperature values outside the predetermined range;

-determining a first analysis parameter (1220) of the first region; and is

-determining a first quality parameter (1230) of a sub-portion to assign the first quality parameter to the first analysis parameter, wherein the first analysis parameter enables to draw conclusions about the quality of the road surface;

wherein the analyzing (1200) is based on learning data comprising at least a learning parameter and at least input data, wherein the input data comprises the first analysis parameter or pattern as the first analysis parameter, and wherein the learning parameter comprises the first quality parameter or a type of error of the paving material at the first sub-portion as the first quality parameter.

18. The device according to claim 17, wherein the computing unit (1400) is based on artificial intelligence and/or is configured to perform a self-learning algorithm.

19. The apparatus of claim 17, wherein the apparatus comprises a thermal profile camera or an asphalt temperature scanner configured to record a thermal profile of the portion when directed to the paving area.

20. The apparatus of claim 17, wherein the apparatus comprises a mobile device and/or a display configured to output information and/or instructions; and/or

Wherein the apparatus comprises a mobile device or control unit receiving (1100) information about the quality parameter from an operator.

21. The device of claim 17, wherein the device comprises a wireless communication module configured to exchange a set comprising the first analysis parameter and the first quality parameter with a server.

22. A paving machine for distributing paving material, comprising the apparatus of claim 17.

Technical Field

Embodiments of the present invention relate to a method for detecting a data set comprising at least a first quality parameter and a first analysis parameter and enabling an assessment of the quality of paving material.

Background

Another embodiment relates to a corresponding apparatus. Another embodiment relates to a paving machine comprising such an apparatus. A general object of embodiments is to improve quality control systems in the field of road construction machines, e.g. asphalt paving machines, based on temperature measurement of new paving material, such as asphalt or asphalt, directly behind the machine or tool (screed).

Quality control in the field of asphalt paving is very important. The processing temperature of the new paving material is an important process variable in road construction which significantly affects the use properties such as stability, layer adhesion and service life of the paved surface. Asphalt paving machines (pavers) typically distribute paving material and pre-compact the surface of the paving material with a screed attached to and pulled by the rear end of the paver. The pavement is then further compacted by rollers. Like other factors such as the environmental and weather conditions during paving, the temperature of the material at different stages of the paving process affects the efficiency and success of the paving operation.

It has long been considered important to treat materials under optimum temperature conditions, but such treatments often require manually controlled measurements on the part of the support and operator. Paving materials are typically obtained at relatively high temperatures in asphalt or asphalt plants. Depending on the distance the supply machine must travel to reach the work site, as well as the traffic and ambient temperature, the bitumen may be cooled to some extent prior to delivery. In addition, the course of paving machines and compacting machines or rollers may vary.

Once the paving material eventually reaches the paving machine or paver, the degree of cooling may vary depending on the temperature of the paving material as it is being transported, environmental factors, and the like. In some instances, paving material may be segregated within the paving machine, and thus there may be relatively cold and relatively hot pockets or accumulations of material within the machine that, once distributed over the work surface, may result in an undesirable, mostly on-time temperature gradient in the paving material. In a typical paving process, paving material is discharged, distributed by a paving machine or paver, then pre-compacted by a screed, and then prepared for further compaction by various compaction machines. During this process, the material temperature may deviate significantly from the expected temperature. Furthermore, the material temperature may be non-uniform from one pavement area to the next due to varying weather conditions or due to undesired segregation or poor mixing.

Due to the importance of the paving temperature of the pavement during the paving process, the measurement of the paving temperature becomes increasingly important. For this purpose, known systems measure the paving temperature behind the paver, in particular behind the screed. There are a number of prior art methods based on thermal imaging.

EP 2789741 a1 describes a road finisher with a thermal imaging device fixed to a part of the machine for recording georeferenced thermal imaging data records of at least one area of the road surface layer. The road finisher comprises a display on which all measurements of the thermal imaging device can be indicated and which provides the possibility of suggesting improved paving parameters to the operator on the basis of the measurements of the thermal imaging device.

EP 2982951 a1 describes a thermal imaging module for a road finisher, which has a temperature measuring device, an imaging device (means) and an evaluation unit. The evaluation unit is configured to determine a deviation area if one or more deviation criteria are fulfilled based on the sensed temperature values.

CN 102691251 a describes a paver asphalt temperature separation infrared detection system and detection method. The system provides real-time monitoring of measured temperatures during asphalt paving. The adverse factors generated during asphalt paving can be obtained using the monitoring results, and a solution can be created at a first time. Adverse factors can be controlled and their effects can be reduced or eliminated altogether, which ensures paving quality.

WO 00/70150 a1 describes a road surface temperature monitoring system having a temperature sensor mounted on the rear end of the paving vehicle so that the entire width of the mat formed can be scanned or imaged. The display device is capable of receiving a plurality of electrical signals from the temperature sensor and generating and displaying a graphical image of the formed pad temperature distribution.

DE 102008058481 a1 describes an asphalt system in which the navigation of the construction vehicle is based on a so-called location temperature model. The system determines where to best employ the compacting vehicle based on the initial assessment and then the measured asphalt temperature.

Furthermore, EP 2990531 a1 and EP 2666908 a1 describe systems and methods of road finisher for determining the cooling behavior of a newly laid asphalt mat.

Some known prior art documents describe proposals to the machine operator to improve machine parameters in the event that the measured temperature of the newly laid paving material is below or above a predetermined temperature. In practice, however, it is not said that these recommendations for the machine operator are always correct, since none of the above-mentioned documents of the prior art deals with the exact root cause of such temperature deviations or analyzes the reality and the real symptoms of thermal separation of the paving material that occurs during the paving process. The correct parameters of the machine are changed primarily based on the experience of the machine operator or paving worker to ensure that paving material will be laid by the paving machine under optimal conditions.

Another prior art document is EP 3456880 a1, which describes a method for controlling the quality of paving material. Here, the thermal profile or thermal pattern is identified and analyzed, for example, based on the orientation of the thermal pattern, in order to draw conclusions from the thermal pattern. However, there are a number of different patterns that enable such conclusions, wherein, according to the prior art, only a limited number of error types are known. Accordingly, there is a need for an improved method.

Disclosure of Invention

It is an object of the present invention to find a concept for enhancing the selection of storage pattern/thermal separation scenarios.

This object is solved by the subject matter of the independent claims.

Embodiments provide a method for detecting a data set comprising at least a first quality parameter and a first analysis parameter and enabling an evaluation of the quality of paving material distributed along a paving area using a paving machine. The method comprises the basic steps of receiving a first thermal profile and analyzing the first thermal profile. The first heat distribution belongs to a portion of a paving area where paving material is distributed; the first thermal profile comprises a plurality of temperature values assigned to respective measurement points, wherein the portion comprises a first sub-portion. Analyzing the first thermal profile has the purpose of detecting a separation point of the paving material within the portion, thereby enabling a determination of the quality of the paving material. The analysis includes:

-determining a first zone of measurement points arranged adjacent to each other and having a temperature value within a predetermined range, wherein the first zone corresponds to the first sub-portion, and wherein the first zone is at least partially surrounded by measurement points having temperature values outside the predetermined range;

-determining a first analysis parameter of the first zone (e.g. orientation of the first zone with respect to the direction of travel or the average temperature); and is

Determining a first quality parameter of the sub-portion (e.g. a parameter describing a type of error of the paving material at the first sub-portion) for assigning the first quality parameter to the first analysis parameter.

According to an embodiment, the method comprises the steps of storing the first analysis parameter together with the first quality parameter; optionally, the method comprises the step of sending the first analysis parameter together with the first quality parameter to store the first analysis parameter and the first quality parameter on the server.

According to a preferred embodiment, the analyzing step comprises a self-learning algorithm and/or is based on artificial intelligence.

According to a preferred embodiment, the method comprises a step of selecting a section of the analysis parameter or one or more possible analysis parameters before performing the analysis step. The set of possible analysis parameters may include:

orientation of the first zone with respect to the direction of travel;

the average temperature of the temperature values within the first zone;

the relative temperature of the temperature values within the first zone when compared to the temperature values belonging to the measurement points surrounding the first zone;

the size of the first region;

temperature deviation within the first zone;

shape of the pattern of the first region;

a second region where there is a measurement point corresponding to a second sub-portion of the portion;

another region of measurement points corresponding to another sub-portion of the portion;

the distance of the first zone to the second zone or the first zone to another zone;

the relative position of the first zone to the second zone or the first zone to another zone; or

A combination of at least two or more group elements.

This pre-selection of analysis parameters helps to train the detection algorithm so that it performs an accurate and fault-tolerant quality analysis (diagnosis).

Embodiments of the present invention are based on the following findings: within the temperature distribution or thermal pattern, one or more analysis parameters may be determined like the shape of the pattern or the temperature deviation from the surroundings and used together with quality information like information about the type of error to determine the data set/learning data. These steps may be performed automatically or semi-automatically, for example, through the use of artificial intelligence. The data set determined in the described manner enables a data basis to be formed, by using which the determination of different situations, for example different data types determined on the basis of the thermal distribution, can be detected. As discussed above, the described methods may be based on self-learning algorithms, and thus artificial intelligence may be used to implement the described methods. This is advantageous for continuously enhancing the data sets and for improving the accuracy, respectively. Another advantage is that the entire method or a large part of it can be performed automatically. Another advantage is that the knowledge generated by the self-learning algorithm may be shared with and/or transferred to other paving systems/control systems of the paving machine.

According to an embodiment, the step of determining the first region of measurement points comprises a step of pattern determination. The pattern may be described by using analysis parameters. In general, the analysis parameter may be (selected from) the group consisting of:

orientation of the first zone with respect to the direction of travel;

the average temperature of the temperature values within the first zone;

the relative temperature of the temperature values within the first zone when compared to the temperature values belonging to the measurement points surrounding the first zone;

the size of the first region;

temperature deviation within the first zone;

shape of the pattern of the first region;

a second region where there is a measurement point corresponding to a second sub-portion of the portion;

another region of measurement points corresponding to another sub-portion of the portion;

the distance of the first zone to the second zone or the first zone to another zone;

the relative position of the first zone to the second zone or the first zone to another zone; or

A combination of at least two or more group elements.

It should be noted that according to an embodiment, the analyzing step is based on learning data comprising at least input data. The input data may comprise first analysis parameters and/or (determined) patterns (e.g. cold spots of a road surface). Alternatively or additionally, the learning parameter may comprise a first quality parameter like an error type of the paving material at the first sub-portion. According to an embodiment, the method includes the step of receiving parameters of the paving machine and/or a configuration of the paving machine. These parameters/configuration information may be used as part of the learning data. The analyzing step is thus based on learning data comprising at least decision parameters defining the type or number of decision nodes, wherein the decision parameters depend on parameters of the road paver or the configuration of the road paver.

According to an embodiment, the method further comprises the step of receiving at least one instruction of a quality parameter or a type of error assigned to the paving material at the first subsection, for example from an operator of the paving machine. This advantageously helps to learn the system based on the experience/knowledge of the operator.

According to an embodiment, the steps of receiving and analyzing are repeated for the same paving area or another paving area, and/or wherein the steps of receiving and analyzing are repeated for another paving machine. This enables to enhance the system to further situations (second quality parameter and second analysis parameter assigned to each other). Here, the second quality parameter/second analysis parameter may refer to a comparable situation (when compared to the first quality parameter/first analysis parameter) or a completely new situation.

According to an embodiment, a plurality of measurement points within a thermal profile are arranged according to a regular grid. According to an embodiment, the method is performed for a plurality of temperature profiles, or wherein the method is performed for a plurality of temperature profiles overlapping each other.

Another embodiment provides an apparatus for detecting a data set comprising at least a first quality parameter and a first analysis parameter. The apparatus includes an interface for receiving the first thermal profile and a computing unit for analyzing the first thermal profile to detect a separation point of paving material within the portion. As discussed above, the analysis processed by the computing unit is performed.

According to an embodiment, the calculation unit is based on artificial intelligence and/or is configured to perform a self-learning algorithm.

It should be noted that, according to embodiments, the training of artificial intelligence algorithms is performed in the cloud or mainframe computer. This means that the image/heat distribution is recorded by the respective construction machine and transmitted as training data to the central unit. The labeling may be performed manually by assigning corresponding quality parameters to the images used as training data, or automatically. The quality parameters form training data together with the images. The improved software may then be sent to the respective work machine, for example, as an update. Based on the new training data, an update algorithm performed on the work machine may be improved for determining problem points based on the thermal profile. The background to the "outsourcing" of the training algorithm is that the process typically requires high computational power.

The above-described embodiments are based on assigned pairs of analysis data and quality data that enable a good determination of problem points. The data set may be enhanced by adding corresponding instructions/work machine parameter variations that enable avoiding the determined problem points. The enhanced data set may also be transmitted to the work machine. Due to the implemented algorithm for assessing road surface quality based on thermal distribution, the validity of recommended/commanded/changed mechanical parameters can be verified. This newly recorded data may be used again as training data. In this case, the command/machine parameters are considered when executing the learning algorithm (e.g., at a large computer).

According to an embodiment, the apparatus comprises a thermal profile camera or an asphalt temperature scanner configured to record a thermal profile of the portion when directed to the paving area. According to an embodiment, the apparatus comprises a mobile device and/or a display configured to output information and/or instructions; additionally or alternatively, the apparatus comprises a mobile device or a control unit receiving information about the quality parameter from an operator. According to an embodiment, the device may include a wireless communication module configured to exchange a set including a first analysis parameter and a first quality parameter with a server.

Another embodiment provides a computer program for performing the above method. Additional embodiments provide a paving machine including the apparatus discussed above.

Before discussing embodiments of the present invention in detail with reference to the accompanying drawings, applications of the above embodiments will be discussed with reference to application examples. The above described embodiments enable detection of a data set comprising at least a first quality parameter and a first analysis parameter. Based on the data set, the quality of the paving material may be evaluated. Application examples describe how this evaluation process can be done. According to a basic application example, a method for controlling the quality of paving material distributed along a paving area using a paving machine is provided.

The method comprises two basic steps: at least one thermal profile pertaining to a portion of a paving area over which paving material is distributed is received from, for example, an infrared camera, and the at least one thermal profile is analyzed to detect a point of separation of the paving material within the portion. The thermal profile comprises a plurality of temperature values assigned to respective measurement points which may be arranged according to a grid. The analysis comprises three sub-steps, namely, determining a first zone of measurement points arranged adjacent to each other and having a temperature value within a predetermined range, wherein the first zone is at least partially surrounded by measurement points having temperature values outside the predetermined range. The next sub-step is to analyze the orientation of the first zone relative to the direction of travel of the paving machine (e.g., to determine that the first zone is oriented either substantially along or perpendicular to the direction of travel). The last sub-step is to assign to the portion a respective type of indication (in particular an error indication, etc.) based on the analysis of the orientation.

According to an enhanced application example, the method comprises the following steps: the temperature deviation within the portion is analyzed, for example, to find a deviation of the temperature value of the first zone from a minimum or maximum temperature value within the portion or from temperature values belonging to measurement points around the zone. According to another application example, the analysis of the thermal distribution comprises a further sub-step of determining a second zone of measurement points (lying next to each other and having temperature values within a further predetermined range, wherein the second zone is at least partially surrounded by measurement points having temperature values outside the further predetermined range, or wherein the second zone adjoins the first zone) and analyzing the orientation of the second zone with respect to the direction of travel.

In general, some application examples may be based on the following findings: there are some criteria like orientation and temperature deviation of the temperature points within the inspected paving area, enabling to parse the thermal data in real time, so that potential causes of separation can be identified, and automatic communication to possible solutions of paving personnel can be given. In order to support the paving personnel, identification patterns (in particular error patterns, etc.) within the thermal profile are identified and assigned to predetermined pattern types or predetermined error types. The assignment of the pattern type or error type enables instructions/prompts to be output to the paving personnel to avoid typical causes of the respective type of pattern or error (thermal separation). An example of such communication to paving personnel may be a real-time warning to a paving operator or paving supervisor in the event of a serious separation problem. The main advantage is that the described system self-analyzes thermal data and gives recommendations without the help of road paver personnel. Thus, independent of the experience of the machine operator or of the paving personnel or of the paving specialist. The application example of the present invention also has the following advantages: the system continuously and uninterruptedly analyzes the measured temperature distribution. When looking at the prior art, there is no guarantee that the machine operator will always look at the display of the operating and display unit and detect every problem that may occur during the paving process. Due to the continuous monitoring, the overall quality of the road to be produced is improved.

Depending on the application example, the analysis of the temperature deviation may have different variants. For example, to analyze the temperature deviation, the sub-step of determining the temperature gradient from one measurement point belonging to the first zone to one measurement point outside the zone may be performed to detect whether the temperature gradient is below or above a predetermined threshold, for example, 25 degrees fahrenheit (150 ° F and 175 ° F) or 50 degrees fahrenheit (150 ° F and 200 ° F) (about 14 ℃ or 28 ℃). In the united states, the thresholds of 25 degrees fahrenheit and 50 degrees fahrenheit are standard variation definitions of "medium" and "severe" separation, and therefore these threshold definitions must be considered examples and may vary according to other paving practices with different materials, widths, and depths. For example, over a 150 foot section, less than 25 degrees fahrenheit (14 degrees celsius) for minimum or no separation; for moderate separation, between 25 degrees fahrenheit and 50 degrees fahrenheit (14 degrees celsius and 28 degrees celsius); and for severe separations, classifications above 50 degrees fahrenheit (28 degrees celsius) are best for many U.S. paving practices. It must be noted, however, that these temperature and distance variations are selected based on studies performed on projects consisting of average U.S. paving practices. Alternatively or additionally, a point-to-point comparison may be performed for two different zones. According to another application example, all temperature values of the portion may be analyzed to determine a lowest temperature and a highest temperature within the portion. This enables a temperature value of the first zone to be compared with a highest or lowest temperature value within the portion.

Depending on the application example, different patterns or error indications may be detected. Here, a distinction can be made between five types (type a to type E). For example, when the first zone is arranged perpendicular or substantially perpendicular to the direction of travel, an error or pattern indication of type a may be detected. Another indicator of type a errors or patterns is the vertical arrangement of the first zone and a temperature change below a predetermined threshold, for example, about 25 degrees fahrenheit or 50 degrees fahrenheit (about 14 degrees celsius or 28 degrees celsius). Errors or patterns of type a result in the lowest to moderate load end separation, which is considered a rougher location in the paving material and/or has a higher air gap. According to an application example, instructions may be output to an operator of the paving machine. The instructions may include one of the following notes:

-ensuring that rows overlap when placed;

-ensuring that the row is placed at an extreme distance not in front of the paver; and is

-ensuring that the pile height in the paver hopper and/or the material transfer vehicle hopper remains consistent and at an acceptable level.

According to an application example, an error or pattern of type B may be indicated when the first zone and the second zone are arranged perpendicular to the direction of travel and when the at least one cold zone is arranged between the first zone and the second zone. The cold zone includes a temperature value that is reduced by at least 50 degrees fahrenheit (at least 28 degrees celsius) when compared to the temperature values of the first and second zones. Another indicator of type B errors or patterns is the above arrangement of the first, second and cold zones in combination with a high temperature change, for example above a predetermined threshold of about 50 degrees fahrenheit (about 28 degrees celsius). Errors or patterns of type B indicate severe load end separation. In the case of errors or patterns of type B, according to a further application example, the method may comprise the additional step of outputting to the operator an instruction having:

-ensuring that rows overlap when placed;

-ensuring that the row is not placed at a distance in front of the laying causing extreme cooling before the laying machine has obtained the laying material;

-ensuring that the paver hopper and/or the material transfer vehicle hopper are not depleted and that the hopper wings are not folded between loads; and is

Ensure proper multipoint loading of the truck.

According to an application example, errors or patterns of type C may be identified and indicated. There is a type C error or pattern when the first and second zones are arranged laterally/along the direction of travel and when one or more cold zones are arranged therebetween. Another additional indicator is a temperature change (seen throughout the section) above a predetermined threshold, for example, about 50 degrees fahrenheit (about 28 degrees celsius). Such errors or patterns of type C result in erroneous separations and may be caused by long stops and excessive screed heating at slow paving speeds. In this case, the method may output instructions to the operator that include:

-minimizing the stopping time; and is

-monitoring the thermal image of the screed plate face at standstill and reducing the screed plate temperature.

According to an application example, an error or pattern of type D may be indicated, wherein the first zone extends laterally along the direction of travel and is centered. An additional indicator is a temperature change of less than about 50 degrees Fahrenheit (about 28 degrees Celsius). In this case, the method may output instructions to the operator that include:

-checking the material movement in the field to determine if further expansion is necessary for consistent flow;

-consider adding mainframe extensions; and is

Ensuring that the recoil pedal is in good condition.

According to an application example, an error or pattern of type E may be indicated when the first zone, the second zone, and the third zone divide the portion into different zones (i.e., at least three vertically arranged zones) along the direction of travel. Here, an additional indicator is a temperature change above a predetermined threshold, such as about 50 degrees fahrenheit (about 28 degrees celsius). Errors or patterns of type E result in load-to-load separation. Here, the method may output instructions to the operator:

ensuring consistent truck transport operations; and is

Ensuring consistent mixing device operation.

Regarding the above method, it should be noted that, according to an application example, a plurality of parts of an area are investigated by the above method. Since the separate analyses are performed consecutively, the sections typically overlap each other, wherein the overlap is due to the travel of the paver in the direction of travel.

Another application example provides a device for carrying out the above-described method, i.e. a device for detecting the quality of paving material distributed along a paving area. The apparatus includes at least an interface for receiving the thermal profile and a calculator for performing an analysis thereof.

According to an application example, the device may additionally comprise a thermal profile camera or an asphalt temperature scanner for reproducing one or more thermal profiles. According to an additional application example, the device may comprise a position sensor like a GPS sensor for continuously measuring the position of the paver and adding position information to the temperature distribution and the advice given by the temperature measuring system of the invention. According to another application example, the device comprises a mobile means or display for outputting information/instructions. The mobile device or display enables the monitoring of the invention. According to another application example, the device comprises a wireless communication module configured to send a real-time alert via a wireless communication link to:

-a mobile device (smartphone and/or smartwatch) of a supervisor of the construction site;

computers, tablets, etc. in a supervision site, remote or close to the construction site;

one or more roller compactor drivers behind the asphalt paver, so that the roller drivers can be informed about problems with the asphalt paving process. The roller driver may then, for example, adjust, modify or optimize the parameters, settings, etc. of the roller compactor; and/or

A remote data server so that, for example, truck drivers and/or personnel of the asphalt mixing plant can be informed about problems with the asphalt material.

For example, the real-time warning CAN be sent via a CAN WLAN gateway module (wireless communication module) which is arranged on the machine and is an interface between a machine communication bus system (e.g. CAN (controller area network) or the like) and a wireless communication system (e.g. WLAN, bluetooth or the like).

In the absence of a separation problem, such a real-time warning may be an output positive message or acknowledgement, for example, when there is no (error) indication for the specified segment or distance of about 150 feet.

According to an embodiment, the method further comprises the step of outputting an instruction to a paving operator based on the first quality parameter. This has the purpose of performing actions or changing parameters of the construction machine based on the quality parameters; additionally or alternatively, the method further comprises the step of (again) determining the first analysis parameter after outputting the instruction or after performing the action or after changing the parameter.

An advantage of an example application is that contractors are provided with a means for using a paving machine to positively affect the quality of paving material distributed along a paving area.

Drawings

Embodiments of the present invention will be discussed later with reference to the accompanying drawings, in which,

FIG. 1a shows a schematic representation of a paving machine, here an asphalt paving machine, comprising an apparatus for controlling the quality of distributed paving material according to a first application example;

FIG. 1b shows a schematic representation of a paving area for illustrating the principle of analyzing a thermal distribution according to a first application example;

figure 2 shows a schematic representation of a control unit belonging to a device for controlling quality;

FIG. 3 shows a schematic illustration of different temperature levels indicating a thermal profile;

fig. 4 to 8 show schematic thermal distributions belonging to a distributed road surface with zones assigned to different pattern types or error types for illustrating the principle of determining different pattern types or error types within a thermal distribution according to an application example;

FIG. 9 shows a schematic thermal profile pertaining to a distributed pavement separation symptom;

FIG. 10 shows a schematic flow diagram of a method of detecting a data set that enables assessment of the quality of paving material;

FIG. 11a shows a schematic block diagram of a calculation unit with which a data set can be determined, according to a basic embodiment; and is

Fig. 11b shows a schematic block diagram of a computing unit comparable to that of fig. 11a but enhanced according to an enhanced embodiment.

Detailed Description

The invention will be discussed later with reference to the drawings, in which the same reference numerals are provided to elements having the same or similar functions so that the descriptions thereof are mutually applicable and interchangeable. Before discussing embodiments of the present invention, the context of the self-learning algorithm will be discussed.

Embodiments of the invention start from a method of controlling the quality of paving material distributed along a paving area using a paving machine and an apparatus for detecting the quality of paving material distributed along a paving area using a paving machine (as disclosed by EP 3456880 a1, quality control systems in the field of road construction machines, for example asphalt paving machines, are based on temperature measurements of new paving material, such as asphalt or asphalt, directly behind the machine or tool (screed)). Embodiments provide a learning mode based on artificial intelligence/self-learning algorithms. In order to interpret the thermal data in real time in the software of the operating and display unit, it is necessary to store the thermal data pattern in the memory of the operating and display unit in order to compare it with the currently measured temperature distribution. These comparison data may be generated by artificial intelligence/self-learning algorithms, enabling the determination or addition of new thermal data patterns (new thermal distribution layouts describing the root cause of thermal separation). It is optionally possible that the machine operator or paving worker or paving specialist must verify the decision made by the system.

In FIG. 1a, a road finisher 10, such as an asphalt paver, is schematically shown. The direction of travel of the road finisher 10 is shown by the arrow F on the ground 120. To distribute paving material over the ground 120 and form the roadway 50, the machine 10 includes a screed 15 attached to a rear end of the machine 10.

Further, machine 10 includes a temperature measuring unit 20 at a rear end thereof, and temperature measuring unit 20 may be, for example, a thermal profile camera or an asphalt temperature scanner. An optional weather station 40 is also arranged in the area of the temperature measuring unit 20, which weather station 40 exemplarily determines the wind speed and the ambient temperature in the area of the road finisher 10. The temperature measurement unit 20 measures the temperature of the surface 110 of the newly laid road surface 50 over the limited road width B in the lateral direction (i.e., transverse to the direction of travel of the road finisher 10) through the outer edges 111 and 112. Thus, the measured values are recorded at measuring points 100 which are schematically shown and which are preferably, but not necessarily, arranged at equal distances d transverse to and/or along the direction of travel of the road finisher 10.

Depending on the exact implementation of the measuring unit 20, the measuring points 100 may be arranged to record a temperature distribution having two dimensions or may only be arranged transversely to the direction of travel F, so that a thermal distribution having two dimensions is recorded during travel along the direction of travel F and is composed of a plurality of measurements along the direction of travel F.

The road finisher 10 of fig. 1a may comprise an operating and display unit 30 electrically connected to the temperature measurement unit 20, the operating and display unit 30 comprising at least a CPU for performing the analysis. As shown in fig. 1a, the operating and display unit 30 may be mounted near the control platform of the paving machine. The operating and display unit 30 may be mounted at any other point of the machine, for example and preferably at the screed 15, so that the paving operator can easily view the display screen. The operating and display unit 30 corresponds to a mobile computer and comprises at least a microcontroller, one or more memory units (RAM, ROM, flash memory …) and one or more input and output devices, such as a touch screen display. As shown in fig. 2, the operating and display unit 30 shows the measured temperature distribution of the surface 110 of the newly laid road surface 50 in a graphical representation on an output device (display screen). Examples of this graphical diagram are shown in fig. 4-9 and described in more detail later.

A machine operator or a paving worker (not shown) can see the measured temperature distribution of the newly paved road 50 on the display screen of the operation and display unit 30. Fig. 2 shows an example of a front side 201 of the operating and display unit 30. In the middle area is a display screen 202, preferably a touch display screen. To the left, right and below the display 202 are a plurality of input keys 205. In the middle area of the display screen 202, a measured temperature profile 204 is graphically shown. Some symbols 203 are shown above, to the right and below the temperature distribution 204. Some symbols 203 show data or information of the current paving process, e.g. a paver speed of 2m/min, a wind speed of 1m/min, a humidity of 41%, etc.

The information shown by unit 30 is very useful for the operator or paving operator, where the information must be continuously monitored. To improve reliability, automatic control of paving quality may be performed when rules exist for interpreting thermal patterns. The method will be discussed with reference to fig. 1 b.

Fig. 1b shows a schematic view of a paving area 47 along which paving material 15 has been distributed using a paving machine traveling in a direction of travel F. As discussed with respect to fig. 1a, the surface 110 of the newly laid road 50 is analyzed using a thermal profile camera or asphalt temperature scanner. The means for determining the thermal distribution captures at least a portion of the area 47, which is marked by reference numeral 49. In the portion 49 where a plurality of measurement points 100 are arranged, a corresponding temperature value can be obtained for each measurement point.

Measurement points arranged adjacent to each other and having a comparable temperature, i.e. a temperature value within a predetermined range, i.e. between 170 and 190 degrees celsius, may be grouped into a common zone.

Here, a first region 51a and a second region 51b are exemplarily shown. The first zone 51a is arranged perpendicularly to the direction of travel F, i.e. transversely with respect to the road surface 15, wherein the zone 51b is arranged along the direction of travel F. Both zones 51a and 51b are typically surrounded by a plurality of measurement points having temperature values outside a predetermined range. Alternatively, the two zones may be arranged adjacent to each other such that only a few measurement points or almost no measurement points outside the predetermined range are arranged between the two zones. Generally, it should be noted that each zone 51a and 51b is formed by a temperature deviation between a local point and the surrounding environment.

The orientation along which the zones 51a and 51b are arranged gives a good indication of the cause of the temperature change. Another indicator of the different causes is the temperature deviation itself. Here, various methods may be performed. For example, the temperature deviation between the zone (e.g., 51a) and the ambient environment may be analyzed. According to another method, a temperature deviation between the two zones 51a and 51b can be detected. Alternatively or additionally, the temperature within one zone of the average temperature within one zone (e.g., within zone 51a) may be compared to the lowest or highest temperature value for portion 49.

With respect to each zone 51a and 51b, it should be noted that typically the temperature deviation within the corresponding zones 51a and 51b is a maximum of 30% or 20%, or according to a preferred embodiment at least 10%, where the percentage is ± from the average temperature within the zones 51a and 51 b.

The above-described automated method for controlling paving quality comprises the steps of determining the corresponding zone 51a and/or 51b and analyzing the orientation of the first zone. Based on the orientation and according to a further example, in combination with the temperature deviation, an assignment of indications (in particular, error indications, etc.) may be performed for the respective portion 49.

Starting from this assignment through the respective pattern type or error type (type a-type E), instructions can be output to the operator that help him to improve the recent situation. The step of outputting the instruction may be performed by using the display unit 30 or the mobile device.

In the following, different temperature profiles will be discussed with respect to fig. 4 to 9, wherein each temperature profile may be assigned to a respective pattern indication or error indication. Fig. 4 to 9 show a new road pavement 50, respectively several different measured temperature profiles of the surface 110 of the newly laid road pavement 50 over the road width B, whereby the different hatching in the shown figures represents the different temperatures shown in the hatching temperature gradient according to fig. 3. Other temperature gradients (e.g., color gradients such as rainbow temperature color gradients or iron temperature color gradients) are known, for example, see reference numeral 206 in fig. 2.

In fig. 3 and also in fig. 4-9, the hot temperature is marked with closer hatching and the cold temperature is marked with lighter hatching, wherein the hot temperature is a temperature of about 356 degrees fahrenheit (about 180 degrees celsius) and the cold temperature is a temperature of about 203 degrees fahrenheit (about 95 degrees celsius).

The distance B 'and the direction F' (shown only in fig. 4 but applicable to all fig. 4 to 9) correspond to the road width B of the newly laid road surface 50 and the direction of travel F of the road finisher 10 (see fig. 1 a). Fig. 4-9 illustrate in detail various thermal separation problems that may occur during the paving process. In the following, the measured temperature distribution, the thermal separation problem, its root cause and possible solutions are described in more detail.

Some portions of the following description relate to so-called heap paving. In such pavers, the hot paving material is not dumped directly into the paver hopper. Instead, the hot paving material is placed as a pile directly on the road in front of the paving machine. Such a pile paver provides a loading conveyor that picks up paving material and loads it into a hopper. Thereafter, the paving material is moved by a second conveyor to a position in front of the transverse screw conveyor and the screed, receiving the hot asphalt paving material directly from the dump truck in the same general manner as a paving machine having a hopper.

Fig. 4 shows a typical temperature distribution of pattern indication or error indication of type a. Here, the level of thermal variation shows minimal to moderate load tip separation, which is generally seen as a coarser location in the bitumen mass, and will have high air voids, making the bitumen more susceptible to early damage due to oxidation, moisture penetration, or resulting in cracking and pitting.

The varying temperature zones (e.g., rows 401 to 404 of fig. 4) have different temperature deviations (levels of thermal variation) transverse to the direction of travel F, F' of the road finisher 10. As can be seen in this example, the overall temperature deviation is limited to about 50 degrees fahrenheit (about 28 degrees celsius).

To improve the situation shown, there are three possible solutions. In the case of windrow paving, it is ensured that the rows overlap when placed. In the case of windrow paving, it is ensured that the rows are not placed at extreme distances in front of the paving machine. Ensuring that the pile height in the paver hopper and/or the material transfer vehicle (material feeder) hopper remains consistent and at an acceptable level.

Fig. 5 shows the level of thermal variation with severe loading end thermal separation, which affects the long-term durability of asphalt pavement concrete structures. The pattern or error is indicated as type B.

The thermal patterns show various temperature zones with high temperature deviations (levels of thermal variation) transverse to the direction of travel F, F' of the road finisher 10 (see rows 501 to 509). Small segments of cold temperature are between some rows (see 510 and 511) and also near the outer edges (see 512 and 513).

Possible solutions are: in the case of windrow paving, it is ensured that the rows overlap when placed. In the case of a windrow paving, it is ensured that the row is not placed at a distance in front of the paving machine before the paving machine picks up the material, causing extreme cooling. Ensure that the paver hopper and/or the material transfer vehicle (material feeder) hopper are not depleted and that the hopper wings are not folded between loads. Proper multi-point loading of the truck is ensured if finally dumped into the paver hopper. The environmental conditions may dictate that a material transfer vehicle (material feeder) is required in the operation for remixing the separated materials.

With respect to fig. 6, type C patterns or errors will be discussed. The thermal image of fig. 6 shows the false separation caused by long stop and excessive screed heating at low paving speeds. Fig. 6 shows two different zones 601 and 602 with various temperature zones. In the region 601, there are several temperature zones with high temperature deviations (thermal variation levels). Sections 603 and 604 are at an intermediate temperature range of about 140 degrees celsius to 150 degrees celsius. Sections 605 and 606 are in a high temperature range of about 170 degrees celsius to 180 degrees celsius and extend transverse to the direction of travel F, F' of the road finisher 10. Sections 607 and 608 are points in the low temperature range of about 95 degrees celsius to 105 degrees celsius. In the area 602, there are mainly two temperature sections 609 and 610 in the high temperature range beside the traveling direction F, F' of the road finisher 10.

Possible solutions are: the stop time is minimized. The thermal image of the screed plate face at rest is monitored and the screed plate temperature is reduced if possible.

With respect to fig. 7, type D patterns or errors will be discussed. Fig. 7 shows a temperature distribution with some thermal striation zones. The image in fig. 7 contains minimal variation due to streaking, but moderate to severe thermal streaking can result in continuous separation of the entire pad. Mainly four areas 701 to 704 are shown in the intermediate temperature range, wherein area 705 shows a linear area beside the direction of travel F, F' of the road finisher 10 also in the intermediate temperature range. The linear region 705 begins at 702 and ends at 703. Further, the region defined by section 706 shows the centerline region in a very high temperature range (up to 180 degrees Celsius).

Possible solutions are: the material movement at the sides was checked to determine if auger spreading was necessary for consistent flow. It is also contemplated to add a main frame extension if the material is moving forward at the end gate. Check for centerline streaking to ensure that the recoil pedal is in good condition. If the kick pedal is in good condition, the flow gate and/or auger height is raised.

Fig. 8 depicts the region tilt for pattern or error type E. Fig. 8 shows a thermal image of a road surface with load-to-load separation. Fig. 8 shows three different zones 801, 802 and 803 with various temperature zones. The areas 801 and 802 show several main temperature sections 804 to 807 within the intermediate and high temperature ranges and transverse to the direction of travel F, F' of the road finisher 10. There is a small boundary between regions 802 and 803 indicating that the paving machine is stopped. Region 803 shows comparable levels of athermal variation to the other two regions 801 and 802. Possible solutions are: ensuring consistent truck transport operation. Ensuring consistent mixing device operation.

For completeness only, fig. 9 illustrates a temperature profile showing minimal to no separation. It can be seen that the level of thermal variation is always below about 25 degrees fahrenheit (about 14 degrees celsius) or less.

As discussed above, the analysis of the temperature pattern is performed automatically, for example by the operating and display unit 30 (see fig. 1) or the CPU of the temperature measurement system. Based on the analysis, the system may output or display instructions for operators and staff using the mobile device. The instructions may include real-time load per time. The real-time alert may include further information such as the underlying cause of the separation of the analyses and/or possible solutions suggested to paving workers. In addition to the alerts given and information provided, the truck driver and/or asphalt mixing plant personnel may adjust, modify, or optimize parameters, settings, etc. of the asphalt production and transportation process. These may, for example, change the temperature of the asphalt mixing process or minimize the number of trucks on the way from the asphalt mixing plant to the job site (to minimize the waiting time of the asphalt trucks in front of the asphalt paver).

According to another example, the output of the instructions may be performed using network technology, such that if the separation problem that occurs is caused by the wrong asphalt mix, the instructions may be given to the asphalt mixing plant.

Although in the above examples the temperature measuring unit has been characterized as a heat distribution camera or asphalt temperature scanner directed at the newly laid asphalt mat behind the asphalt paver, it should be noted that other measuring principles (such as using a plurality of sensors) are also possible if the measuring principles enable the heat distribution to be captured.

According to another embodiment, a so-called learning mode may be used. The purpose of this learning mode (also referred to as a teach mode) is to add new thermal data patterns (new thermal profiles describing thermally separated resources) to or generate new data sets to already available data sets. According to a preferred implementation, the learning mode is performed automatically or semi-automatically, for example by using artificial intelligence or in general self-learning algorithms.

Next, with reference to fig. 10, a corresponding method 1000 will be discussed.

Fig. 10 shows a method 1000 of detecting a data set comprising at least a first quality parameter (e.g. describing the type of error) and a first analysis parameter (e.g. a specific thermal pattern or a parameter which generally enables conclusions to be drawn about the quality of the road surface, wherein the analysis parameter is determinable, for example, within an image/thermal distribution).

The first quality parameter and the first analysis parameter together describe an identifiable situation, for example comprising a typical error and typical parameters for identifying said error. The first quality parameter or generally one or more quality parameters may be information about the error/type of error or may also be a measured value, for example the evenness of the road surface or the like. The analysis parameters describe parameters that characterize a first region (e.g., a spot) within which errors are detectable. The zone is defined as measurement points arranged adjacent to each other and having all temperature values within a predetermined range (e.g. at least 10% higher than the temperature values of the measurement points partially enclosing the first zone). When the partial distribution is seen, the first zone corresponds to a first sub-portion of a portion of the paving area. The first sub-portion is the region where the first quality parameter is determined. The two parameters (first quality parameter and first analysis parameter) together form a data set on the basis of which the quality of the paving material can be evaluated by using the system described above. The (semi-automatic/automatic) determination of the data set may be done using a method 1000 comprising two basic steps 1100 and 1200.

In step 1100, a first thermal profile of a portion of a paving area having paving material distributed thereon is received. The portion includes a first sub-portion. It is assumed that in said first subsection there is a locally limited quality situation, e.g. an error. The temperature values of the points of the first sub-part are comprised in a first thermal distribution, wherein a first region of measurement points is assigned to the first sub-part. In other words, the first thermal profile comprises a plurality of temperature values assigned to the respective measurement points of the portion, wherein the selection of temperature values refers to the measurement points of the first zone and belong to the first sub-portion. Of course, in addition to the first sub-portion (the portion of interest/to be analyzed), the entire portion may include other/second sub-portions that may be assigned to other/second zones within the thermal profile.

The next basic step is to analyze the first thermal profile. This step is labeled in fig. 10 by reference numeral 1200. The analysis step 1200 includes three sub-steps 1210, 1220 and 1230.

In step 1210, first regions of measurement points that are adjacent to each other and have a temperature range within a predetermined range are determined. The first zone is at least partially surrounded by a measurement point having a temperature outside a predetermined range, for example forming a second zone or a continuous area. For example, the first zone may be interpreted as a spot having a very high temperature value, i.e. a so-called hot spot within the overall temperature distribution.

In a next step 1220, a first feature analysis parameter for the first region is determined. For example, when compared to temperature values belonging to measurement points around the first zone, the average temperature of the temperature values within the first zone or the relative temperature of the temperature values within the first zone may be determined as the first feature analysis parameter. Additionally or alternatively, the geometry of the first zone (e.g., orientation relative to the direction of travel of the first zone or size or shape of the first zone) may be determined as a feature analysis parameter. One or more of these analysis parameters are used to enable identification of zones where comparable identifiable situations/comparable errors are assumed to be likely to occur. Preferably, the first region is described in detail using more than one or all of the analysis parameters.

For the sub-portion corresponding to the first region within the first thermal profile, a quality parameter, e.g. uniformity or error type, is determined in step 1220. According to an embodiment, the quality parameter may be determined manually (i.e. by an operator). Possible variants are manual measurements such as uniformity of separation or visual inspection. The result of this step 1220 is that the respective first analysis parameter or the respective first set of analysis parameters is assigned to the respective quality parameter.

In step 1230, the combination of analysis parameters and quality parameters is stored, for example, locally or on a server.

According to an embodiment, it is advantageous to repeat these steps 1100 and 1200 for a plurality of comparable cases. For example, if the same quality parameter, e.g., the same error type, is found at different sub-portions (see step 1330), the self-running algorithm may find similarities within the analysis parameters in order to train itself to determine the relevant error type. For example, if it is determined that corresponding regions belonging to the same error type have the same orientation, the algorithm may be trained in such a way that the orientation indicates the corresponding error type. Alternatively, a temperature deviation between temperature values within a point or a temperature deviation between average temperatures of points compared to average temperatures outside a point may be considered. For example, when the point temperature deviates from the surroundings by more than 10 degrees and when the orientation has a certain angle (e.g., 90 °) with respect to the direction of travel, the algorithm may find that a corresponding error belonging to a corresponding error type generally occurs. If both analysis parameters can be determined each time a respective error is found, it is possible to deduce by using this method that the two analysis parameters of the plurality of analysis parameters are correlated and which respective threshold value indicates that a certain condition exists. Since there are multiple possible analysis parameters and the correlation of a single parameter or a combination of single parameters with respect to the type of error can be trained by using training data. The parameters may be one or more of the group:

orientation of the first zone with respect to the direction of travel;

the average temperature of the temperature values within the first zone;

the relative temperature of the temperature values within the first zone when compared to the temperature values belonging to measurement points around the first zone;

the size of the first region;

temperature deviation within the first zone;

shape of the pattern of the first region;

a second region where there is a measurement point corresponding to a second sub-portion of the portion;

another region of measurement points corresponding to another sub-portion of the portion;

the distance of the first zone to the second zone or the first zone to another zone; and

the relative position of the first zone to the second zone or the first zone to another zone.

It should be noted that different analysis parameters may be correlated for different error types.

In summary, the two steps 1100 and 1200 are preferably repeated for different cases mapping comparable error types to analysis parameters in order to find out which analysis parameters and which values of said analysis parameters indicate the respective case/error. Furthermore, the two steps 1100 and 1200 are preferably repeated for different cases mapping different error types to analysis parameters in order to find out which analysis parameters indicate different cases/errors. These error types or generally quality parameters together with one or more analysis parameters form so-called training data, enabling the algorithm to be trained.

According to a further embodiment, the data set for the respective case can be stored as a solution, such as paving machine parameters to be changed in order to avoid the determined error.

Starting from this central process of determining the analysis parameters, it is possible to draw conclusions about the quality of the road surface (driving direction, orientation of the points, temperature differences between different zones). This process can be enhanced as follows. In general, the method 1000 described above enables extraction of analysis parameters from a temperature distribution or another image (e.g., a color image of a road surface). These are unusual or undesirable parameters. During the diagnostic phase, quality parameters and optionally potential measurements may be determined. After this measurement is carried out, the image of the road surface changes, for example becomes uniform. Thus, the respective analysis parameter will be below the respective threshold value and the quality of the road surface is within specification. The process can be described as follows:

-determining or using an analysis parameter;

-analyzing the image, e.g. a thermography with respect to an analysis parameter;

-performing a diagnosis to determine a quality problem of the road surface, i.e. to determine a quality parameter;

-providing a message on potential measurements to solve quality problems;

-carrying out the measurement;

-as a result of the implementation, changing the respective analysis parameter;

-receiving a new image and analyzing the analysis parameters again.

The determination of the analysis parameters or the verification of the analysis parameters is important for the analysis performed by artificial intelligence.

Its context is independent of the analysis parameters, and a correspondingly trained artificial intelligence image recognition algorithm can be used to analyze the images and extract the diagnosis. In this case, the training data set only comprises images and corresponding quality parameters enabling a diagnosis to be performed. This process is called "tagging" and may be supported by a human, for example, by selecting the analysis parameters on which tagging should be performed. In the absence of human support, it may happen that the difference between different quality situations is misleading. For example, an artificial intelligence algorithm trained to perform differentiation between images of a husky and a wolf may be based on erroneous parameters, such as background, since huskies are typically recorded in winter landscapes. When such wrong parameters are automatically determined by an artificial intelligence algorithm, most decisions can be correct, but put in the parameters that are completely wrong, because such an algorithm cannot distinguish between husky and wolf, and more so, between animals in a winter landscape or animals in another landscape. Thus, according to an embodiment, the analysis parameters are pre-selected.

With respect to the proposed measurements/instructions to be output to the paving operator, it should be noted that, depending on the implementation, the validity of these measurements may be proven. For example, the same quality parameters of the respective images taken after the measurements have been carried out can be determined to prove whether the action taken was successful. According to an embodiment, information about the respective analysis parameters may be output to a central unit, e.g. performing training of the algorithm, in order to evaluate the proposed measurements. According to a further embodiment, the changed parameters of the work machine/road surface can be monitored and fed back to the training algorithm together with the analysis parameters and optionally with the quality parameters in order to monitor which parameters lead to which result in which case. The system so implemented may continuously learn the determination of new problem points and continuously improve the effectiveness of the selected solution (e.g., change machine parameters).

The method 1000 may preferably be accomplished by forming a neural circuit of artificial intelligence. Here, a plurality of analysis parameters determined for each case (point) are used as input parameters for the neural circuit, wherein the nodes are determined based on the correlation of the respective analysis parameters in relation to comparable quality parameters. Fig. 11a and 11b show such a black box of an artificial intelligence based computing unit.

Fig. 11a shows a calculation unit 1400 receiving at least a first thermal profile IR, for example from an interface 1410 connected to an infrared camera 1420 (general temperature sensor). The temperature profile may be for includingAn infrared image IR of a corresponding portion of the road surface of at least the first subsection. This first sub-portion is in parallel with the temperature analysis, for example by an operator. Based on this analysis, a quality parameter QP is generated and forwarded to the calculation unit 1400. These two parameters IR and QP are input parameters. These input parameters IR₁、IR₂、IR₃、IR_nAnd QP₁、QP₂、QP₃And QP_nPreferably received for a plurality of situations, e.g. a plurality of sub-parts or portions. These portions may be determined using the same paving area or different paving areas using the same or different paving machines.

Based on at least one pair of parameters IR₁/OQ₁Or IR₂/OQ₂Or IR₃/OQ₃Respective data sets are generated that are assigned to each other. Let QP be assumed₁And QP1₃Describing comparable quality parameters, e.g. the same error type, the similarity between the analysis parameters, i.e. the IR, can be determined₁And IR₃The similarity between them. Each pair of IR₁And QP₁And IR₃And QP₃Corresponding data sets are formed, for example, data set case 1 and data set case 1'. Both data sets may be used to describe the respective error type. By having a plurality of such data sets, all analysis parameters and the relevant analysis parameters in the corresponding threshold/threshold range (greater than 120 degrees or in the range of 115 degrees to 130 degrees) can be determined. The plurality of data sets "sit.1" improves the determination efficiency and the determination accuracy of the respective cases. Of course, depending on the embodiment, different situations may be determined, e.g. belonging to different quality parameters. Here, the data set IR₂And QP₂A data set is generated that describes different situations, here case 2 (data set "sit.2").

It should be noted that data set sit.2 and data set sit.1' are marked with cross-hatching because multiple pairs of ER are analyzed according to the basic implementation_nAnd QP_nBut merely optional.

Fig. 11b shows another configuration of the computing unit 1400. The calculation unit receives the same input parameters IR and OP to determine their respective data sets. However, the calculation unit 1400 according to the embodiment of fig. 11b further enables to receive decision parameters as part of the learning data. The decision parameter may be a parameter determined, for example, by using a different sensor. For example, the decision parameters may depend on the operational parameters of the paving machine or the configuration of the paving machine. For example, the width of the screen or the temperature of the paving material may have an effect. The decision parameter may be used as an input parameter or may be used to adapt the nodes, e.g. the number of nodes to be combined or an analysis parameter.

According to an embodiment, the determination algorithm determined by using the training data has a plurality of decision layers. For example, in a first decision layer, it may be determined whether the detection zone has an orientation. In the next decision, it can be determined whether the orientation is perpendicular or parallel to the direction of travel. Within the next decision layer, it can be determined whether the zone is a hot zone or a cold zone. By using these decisions, the situation can be clearly assigned to the quality parameter. However, other analysis parameters or another sequence may have the same result. The self-learning algorithm enables to find out which parameters and which order lead to a high determination efficiency and accuracy. The context of which is the order of the corresponding decisions may have an impact; for example, the error that results in a separation point parallel to the direction of travel may have a temperature value of a low temperature level or a moderate temperature level. The temperature from one end to the other end along the direction of travel differs due to the different duration of cooling at different locations within the point along the direction of travel. Thus, the determination of longitudinal points along the direction of travel may make some analysis of the absolute temperature values, for example, obsolete. Therefore, it may be beneficial to determine the orientation (i.e., one of the first decision layers) first. By using multiple training data, cross-references between corresponding analysis parameters can be found. Preferably, this is done by an artificial intelligence method that enables the benefits of large data (large amounts of training data) to be used.

Although some aspects have been described in the context of a device, it is clear that these aspects also represent a description of the corresponding method, wherein the blocks or means correspond to method steps or features of method steps. Similarly, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding device. Some or all of the method steps may be performed by (or using) a hardware device, such as, for example, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some one or more of the most important method steps may be performed by such a device.

Embodiments of the invention may be implemented in hardware or software, depending on certain implementation requirements. The implementation can be performed using a digital storage medium (e.g. a floppy disk, a DVD, a blu-ray, a CD, a ROM, a PROM, an EPROM, an EEPROM or a flash memory) having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Accordingly, the digital storage medium may be computer-readable.

Some embodiments according to the invention comprise a data carrier with electronically readable control signals capable of cooperating with a programmable computer system such that one of the methods described herein is performed.

In general, embodiments of the invention can be implemented as a computer program product having a program code operable to perform one of the methods when the computer program product runs on a computer. The program code may be stored, for example, on a machine-readable carrier.

Other embodiments include a computer program stored on a machine-readable carrier for performing one of the methods described herein.

In other words, an embodiment of the method of the invention is thus a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

Thus, a further embodiment of the inventive method is a data carrier (or digital storage medium, or computer readable medium) comprising a computer program recorded thereon for performing one of the methods described herein. Data carriers, digital storage media or recording media are usually tangible and/or non-transitory.

A further embodiment of the inventive method is thus a data stream or a signal sequence representing a computer program for performing one of the methods described herein. The data stream or signal sequence may for example be arranged to be transmitted via a data communication connection, e.g. via the internet.

Further embodiments include a processing device, e.g., a computer or a programmable logic device, configured or adapted to perform one of the methods described herein.

Further embodiments include a computer having installed thereon a computer program for performing one of the methods described herein.

Further embodiments according to the present invention include a device or system configured to transmit (e.g., electronically or optically) a computer program for performing one of the methods described herein to a receiver. The receiver may be, for example, a computer, a mobile device, a storage device, etc. The device or system may for example comprise a file server for transmitting the computer program to the receiver.

In some embodiments, a programmable logic device (e.g., a field programmable gate array) may be used to perform some or all of the functions of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor to perform one of the methods described herein. In general, the methods are preferably performed by any hardware device.

The above-described embodiments are merely illustrative of the principles of the present invention. It is to be understood that modifications and variations of the arrangements and details described herein will be apparent to others skilled in the art. Therefore, its purpose is to be limited only by the scope of the following patent claims and not by the specific details presented by the description and explanation of the embodiments herein.