阅读说明：本技术 用于飞行器的障碍物探测方法、装置及系统 (Obstacle detection method, device and system for aircraft ) 是由 沙承贤 于 2020-11-03 设计创作，主要内容包括：本公开涉及用于飞行器的障碍物探测方法、装置及系统。障碍物探测方法包括：获取安装于飞行器上的多个雷达传感器在当前时刻探测位于指定方向上的多个待探测区域得到的探测结果,一个雷达传感器发射的波束形成一个待探测区域,每个探测结果表征对应的待探测区域是否存在障碍物,所述多个待探测区域交叠构成包括多个栅格的占用栅格地图,所述多个栅格包括一个或多个重叠区域；根据所获取的与指定的栅格所在的待探测区域对应的至少一个雷达传感器的探测结果,利用正演模型,确定所述指定的栅格被占用的实际占用概率和未被占用的实际非占用概率；对于所述指定的栅格,根据对应的实际占用概率和实际非占用概率,确定所述指定的栅格是否存在障碍物。(The present disclosure relates to obstacle detection methods, devices, and systems for aircraft. The obstacle detection method includes: the method comprises the steps that detection results obtained when a plurality of radar sensors installed on an aircraft detect a plurality of to-be-detected areas located in a specified direction at the current moment are obtained, a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas; determining the occupied actual occupation probability and the unoccupied actual non-occupation probability of the designated grid by utilizing a forward model according to the acquired detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located; and determining whether the specified grid has an obstacle or not according to the corresponding actual occupation probability and actual non-occupation probability.)

1. An obstacle detection method for an aircraft, comprising:

acquiring detection results obtained by a plurality of radar sensors installed on the aircraft at the current moment for detecting a plurality of to-be-detected areas in a specified direction, wherein a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas;

determining the occupied actual occupation probability and the unoccupied actual non-occupation probability of the designated grid by utilizing a forward model according to the acquired detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located;

and determining whether the specified grid has an obstacle or not according to the corresponding actual occupation probability and actual non-occupation probability.

2. The obstacle detection method for an aircraft according to claim 1, wherein determining, using a forward model, an actual occupancy probability that a specified grid is occupied based on the acquired detection results of at least one radar sensor corresponding to a region to be detected in which the specified grid is located comprises:

determining the actual occupation probability of the appointed grid occupied by utilizing a forward modeling according to the acquired detection result of at least one radar sensor corresponding to the area to be detected where the appointed grid is located and a plurality of preset conditional occupation probabilities of the at least one radar sensor, wherein,

the conditional occupancy probability of each of the at least one radar sensor comprises a first conditional occupancy probability and a second conditional occupancy probability,

the first conditional occupation probability represents the probability that the detection result of the radar sensor is that an obstacle exists in the area to be detected in which the designated grid is located under the condition that the designated grid is occupied,

the second conditional occupation probability represents a probability that the detection result of the radar sensor is that no obstacle exists in the area to be detected where the designated grid is located under the condition that the designated grid is occupied, and the first conditional occupation probability is greater than the second conditional occupation probability.

3. The obstacle detection method for an aircraft according to claim 2, wherein determining, using a forward model, an actual occupancy probability that the designated grid is occupied based on the acquired detection results of at least one radar sensor corresponding to the area to be detected in which the designated grid is located and a plurality of preset conditional occupancy probabilities of the at least one radar sensor comprises:

determining an occupancy value of the designated occupancy grid, wherein the occupancy value is a ratio of a priori occupancy probability and a priori non-occupancy probability of the designated occupancy grid, and the priori occupancy probability is smaller than the priori non-occupancy probability;

for each radar sensor in the at least one radar sensor, selecting a conditional occupation probability corresponding to the obtained detection result from the conditional occupation probabilities of the radar sensor according to the obtained detection result;

determining an actual occupancy probability that the designated grid is occupied according to the selected conditional occupancy probability of the at least one radar sensor and the occupancy value.

4. The obstacle detection method for an aircraft according to claim 3, wherein, for each of the at least one radar sensor,

under the condition that the obtained detection result indicates that the obstacle exists, the selected corresponding conditional occupation probability is a first conditional occupation probability;

and under the condition that the obtained detection result is that no obstacle exists, the selected corresponding conditional occupation probability is the second conditional occupation probability.

5. The obstacle detection method for an aircraft according to claim 2, wherein determining, using a forward model, an actual non-occupancy probability that a specified grid is unoccupied, based on the acquired detection results of at least one radar sensor corresponding to the area to be detected in which the specified grid is located, comprises:

determining the actual unoccupied probability of each grid which is not occupied by utilizing a forward modeling model according to the acquired detection result of at least one radar sensor corresponding to the area to be detected where the specified grid is located and a plurality of conditional unoccupied probabilities of the at least one radar sensor, wherein,

the conditional non-occupancy probability for each of the at least one radar sensor comprises a first conditional non-occupancy probability and a second conditional non-occupancy probability,

the first condition non-occupation probability represents the probability that the detection result of the radar sensor is that no obstacle exists in the area to be detected where the specified grid is located under the condition that the specified grid is not occupied;

the second condition non-occupation probability represents that under the condition that the specified grid is not occupied, the detection result of the radar sensor is the probability that an obstacle exists in the area to be detected where the specified grid is located, the first condition non-occupation probability is greater than the second condition non-occupation probability, the first condition non-occupation probability is greater than the first condition occupation probability, and the second condition non-occupation probability is smaller than the second condition occupation probability.

6. The obstacle detection method for an aircraft according to claim 5, wherein determining, using a forward model, an actual non-occupancy probability that the designated grid is unoccupied, based on the acquired detection results of the at least one radar sensor corresponding to the area to be detected in which the designated grid is located and a plurality of preset conditional non-occupancy probabilities of the at least one radar sensor, comprises:

for each radar sensor in the at least one radar sensor, selecting a conditional non-occupation probability corresponding to the obtained detection result from the conditional non-occupation probabilities of the radar sensor according to the obtained detection result;

determining an actual non-occupancy probability that the designated grid is unoccupied based on the selected conditional non-occupancy probability of the at least one radar sensor.

7. The obstacle detection method for an aircraft according to claim 6, wherein, for each of the at least one radar sensor,

under the condition that the obtained detection result indicates that no obstacle exists, the selected corresponding conditional non-occupation probability is a first conditional non-occupation probability;

and in the case that the acquired detection result is that the obstacle exists, the selected corresponding conditional non-occupation probability is the second conditional non-occupation probability.

8. The obstacle detection method for an aircraft according to claim 1, wherein determining whether an obstacle exists for the specified grid includes:

comparing the magnitude relation of the logarithm of the actual occupation probability and the logarithm of the actual non-occupation probability;

and determining whether the specified grid has an obstacle or not according to the comparison result.

9. The obstacle detection method for an aircraft according to claim 8, wherein a base of the logarithm is greater than 1, and determining whether an obstacle exists in the designated grid according to a result of the comparison includes:

determining that an obstacle exists in the designated grid under the condition that the logarithm of the actual occupation probability is larger than the logarithm of the actual non-occupation probability as a comparison result;

determining that the designated grid is free of obstacles if the logarithm of the actual probability of occupancy is less than the logarithm of the actual probability of non-occupancy as a result of the comparison.

10. An obstacle detection method for an aircraft according to claim 1, wherein the radar sensor is an ultrasonic radar sensor or a millimeter wave radar sensor.

11. The obstacle detection method for an aircraft according to claim 1, wherein beam directions of beams emitted by the plurality of radar sensors are the same.

12. The obstacle detection method for an aircraft according to claim 1, wherein the plurality of radar sensors are two radar sensors, the specified direction is a flight direction of the aircraft, the overlap region is one that is located directly in front of the aircraft and that enables the aircraft to fly therethrough, and the specified grid is an overlap region.

13. The obstruction detection method for an aircraft according to claim 1, further comprising:

establishing a correspondence relationship between detection results of the plurality of radar sensors and a result of the determined presence or absence of an obstacle in the designated grid after determining the presence or absence of an obstacle in the designated grid;

and under the condition that detection results obtained by detecting the plurality of areas to be detected by the plurality of radar sensors at subsequent time are obtained, determining whether the specified grid has an obstacle or not according to the established corresponding relation.

14. An obstacle detecting device for an aircraft, comprising:

the acquisition module is configured to acquire detection results obtained by detecting a plurality of to-be-detected areas located in a specified direction at the current moment by a plurality of radar sensors installed on the aircraft, wherein a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas;

the first determination module is configured to determine an actual occupied probability and an actual unoccupied probability of the specified grid by utilizing a forward model according to the acquired detection result of at least one radar sensor corresponding to the to-be-detected region where the specified grid is located;

a second determination module configured to determine, for the designated grid, whether an obstacle exists in the designated grid according to the corresponding actual occupancy probability and actual non-occupancy probability.

15. An obstacle detecting device for an aircraft, comprising:

a memory; and

a processor coupled to the memory, the processor configured to perform the method for obstacle detection for an aircraft of any of claims 1-13 based on instructions stored in the memory.

16. An obstacle detection system for an aircraft, comprising:

the system comprises a plurality of radar sensors, a plurality of sensors and a plurality of communication modules, wherein the radar sensors are arranged on the aircraft and configured to detect a plurality of to-be-detected areas located in a specified direction at the current moment to obtain detection results, a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas; and

an obstacle detecting device for an aircraft according to any one of claims 14 to 15.

17. A computer-storable medium having stored thereon computer program instructions which, when executed by a processor, implement the method for obstacle detection for an aircraft according to any one of claims 1 to 13.

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a method, an apparatus, and a system for detecting an obstacle for an aircraft, and a computer-readable storage medium.

Background art,

With the rapid development of social economy, the unmanned aerial vehicle is applied to industries such as logistics and the like, and becomes an important research direction for promoting the development of the unmanned aerial vehicle industry. In the unmanned aerial vehicle field, unmanned aerial vehicle keeps away barrier function and belongs to the crucial one in the unmanned aerial vehicle technique. The obstacle avoidance of the unmanned aerial vehicle has the premise that accurate obstacle detection is carried out, and the obstacle detection technology of the unmanned aerial vehicle is developed at the right moment.

In the correlation technique, a binocular camera and a radar sensor are fused to detect obstacles in the field of unmanned aerial vehicles.

Disclosure of Invention

The inventor thinks that: among the correlation technique, binocular camera receives the influence of environmental conditions such as sleet fog easily, and under adverse environmental condition, the accuracy of barrier detection is relatively poor, and in addition, binocular camera's detection distance is than shorter, and the flying speed of aircraft is usually very fast, and when binocular camera detected the barrier, the aircraft had been very close apart from the barrier, and the efficiency that the barrier was kept away in flight to the aircraft is lower.

To above-mentioned technical problem, this disclosure provides a solution, can improve the barrier detection's of aircraft accuracy nature, and then can improve the efficiency that the aircraft carried out the flight and keeps away the barrier.

According to a first aspect of the present disclosure, there is provided an obstacle detection method for an aircraft, comprising: acquiring detection results obtained by a plurality of radar sensors installed on the aircraft at the current moment for detecting a plurality of to-be-detected areas in a specified direction, wherein a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas; determining the occupied actual occupation probability and the unoccupied actual non-occupation probability of the designated grid by utilizing a forward model according to the acquired detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located; and determining whether the specified grid has an obstacle or not according to the corresponding actual occupation probability and actual non-occupation probability.

In some embodiments, determining, by using a forward modeling, an actual occupation probability that the designated grid is occupied according to the obtained detection result of the at least one radar sensor corresponding to the to-be-detected region where the designated grid is located includes: determining an actual occupation probability of the designated grid by using a forward model according to the acquired detection result of at least one radar sensor corresponding to the to-be-detected area where the designated grid is located and a plurality of preset conditional occupation probabilities of the at least one radar sensor, wherein the conditional occupation probability of each radar sensor in the at least one radar sensor comprises a first conditional occupation probability and a second conditional occupation probability, the first conditional occupation probability represents that the detection result of the radar sensor is the probability that an obstacle exists in the to-be-detected area where the designated grid is located under the condition that the designated grid is occupied, and the second conditional occupation probability represents that the detection result of the radar sensor is the probability that an obstacle does not exist in the to-be-detected area where the designated grid is located under the condition that the designated grid is occupied, the first conditional occupancy probability is greater than the second conditional occupancy probability.

In some embodiments, determining, by using a forward modeling, an actual occupation probability that a designated grid is occupied according to an acquired detection result of at least one radar sensor corresponding to a to-be-detected region where the designated grid is located and a preset plurality of conditional occupation probabilities of the at least one radar sensor includes: determining an occupancy value of the designated occupancy grid, wherein the occupancy value is a ratio of a priori occupancy probability and a priori non-occupancy probability of the designated occupancy grid, and the priori occupancy probability is smaller than the priori non-occupancy probability; for each radar sensor in the at least one radar sensor, selecting a conditional occupation probability corresponding to the obtained detection result from the conditional occupation probabilities of the radar sensor according to the obtained detection result; determining an actual occupancy probability that the designated grid is occupied according to the selected conditional occupancy probability of the at least one radar sensor and the occupancy value.

In some embodiments, for each of the at least one radar sensor, in the case that the obtained detection result is that an obstacle is present, the selected corresponding conditional occupancy probability is a first conditional occupancy probability; and under the condition that the obtained detection result is that no obstacle exists, the selected corresponding conditional occupation probability is the second conditional occupation probability.

In some embodiments, determining, by using a forward modeling, an actual non-occupation probability that the designated grid is not occupied according to the obtained detection result of the at least one radar sensor corresponding to the to-be-detected region where the designated grid is located includes: determining an actual unoccupied probability of each grid which is unoccupied by using a forward modeling according to the acquired detection result of at least one radar sensor corresponding to the to-be-detected area where the specified grid is located and a plurality of conditional unoccupied probabilities of the at least one radar sensor, wherein the conditional unoccupied probability of each radar sensor in the at least one radar sensor comprises a first conditional unoccupied probability and a second conditional unoccupied probability, and the first conditional unoccupied probability represents that the detection result of the radar sensor is the probability that no obstacle exists in the to-be-detected area where the specified grid is located under the condition that the specified grid is unoccupied; the second condition non-occupation probability represents that under the condition that the specified grid is not occupied, the detection result of the radar sensor is the probability that an obstacle exists in the area to be detected where the specified grid is located, the first condition non-occupation probability is greater than the second condition non-occupation probability, the first condition non-occupation probability is greater than the first condition occupation probability, and the second condition non-occupation probability is smaller than the second condition occupation probability.

In some embodiments, determining, by using a forward modeling, an actual non-occupation probability that the designated grid is not occupied according to the obtained detection result of the at least one radar sensor corresponding to the to-be-detected region where the designated grid is located and a plurality of preset conditional non-occupation probabilities of the at least one radar sensor includes: for each radar sensor in the at least one radar sensor, selecting a conditional non-occupation probability corresponding to the obtained detection result from the conditional non-occupation probabilities of the radar sensor according to the obtained detection result; determining an actual non-occupancy probability that the designated grid is unoccupied based on the selected conditional non-occupancy probability of the at least one radar sensor.

In some embodiments, for each of the at least one radar sensor, in the case that the obtained detection result is that no obstacle is present, the selected corresponding conditional non-occupancy probability is a first conditional non-occupancy probability; and in the case that the acquired detection result is that the obstacle exists, the selected corresponding conditional non-occupation probability is the second conditional non-occupation probability.

In some embodiments, determining whether an obstacle is present in the designated grid comprises: comparing the magnitude relation of the logarithm of the actual occupation probability and the logarithm of the actual non-occupation probability; and determining whether the specified grid has an obstacle or not according to the comparison result.

In some embodiments, the base of the logarithm is greater than 1, and determining whether an obstacle is present in the designated grid based on the comparison comprises: determining that an obstacle exists in the designated grid under the condition that the logarithm of the actual occupation probability is larger than the logarithm of the actual non-occupation probability as a comparison result; determining that the designated grid is free of obstacles if the logarithm of the actual probability of occupancy is less than the logarithm of the actual probability of non-occupancy as a result of the comparison.

In some embodiments, the radar sensor is an ultrasonic radar sensor or a millimeter wave radar sensor.

In some embodiments, the beam directions of the beams transmitted by the plurality of radar sensors are the same.

In some embodiments, the plurality of radar sensors are two radar sensors, the designated direction is a flight direction of the aircraft, the overlap region is one, the overlap region is located directly in front of the aircraft and enables the aircraft to fly through the overlap region, and the designated grid is the overlap region.

In some embodiments, the obstacle detection method for an aircraft further comprises: establishing a correspondence relationship between detection results of the plurality of radar sensors and a result of the determined presence or absence of an obstacle in the designated grid after determining the presence or absence of an obstacle in the designated grid; and under the condition that detection results obtained by detecting the plurality of areas to be detected by the plurality of radar sensors at subsequent time are obtained, determining whether the specified grid has an obstacle or not according to the established corresponding relation.

According to a second aspect of the present disclosure, there is provided an obstacle detecting device for an aircraft, comprising: the acquisition module is configured to acquire detection results obtained by detecting a plurality of to-be-detected areas located in a specified direction at the current moment by a plurality of radar sensors installed on the aircraft, wherein a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas; the first determination module is configured to determine an actual occupied probability and an actual unoccupied probability of the specified grid by utilizing a forward model according to the acquired detection result of at least one radar sensor corresponding to the to-be-detected region where the specified grid is located; a second determination module configured to determine, for the designated grid, whether an obstacle exists in the designated grid according to the corresponding actual occupancy probability and actual non-occupancy probability.

According to a third aspect of the present disclosure, there is provided an obstacle detecting device for an aircraft, comprising: a memory; and a processor coupled to the memory, the processor configured to perform the method for obstacle detection for an aircraft of any of the embodiments described above based on instructions stored in the memory.

According to a fourth aspect of the present disclosure, there is provided an obstacle detection system for an aircraft, comprising: the system comprises a plurality of radar sensors, a plurality of sensors and a plurality of communication modules, wherein the radar sensors are arranged on the aircraft and configured to detect a plurality of to-be-detected areas located in a specified direction at the current moment to obtain detection results, a beam emitted by one radar sensor forms one to-be-detected area, each detection result represents whether an obstacle exists in the corresponding to-be-detected area, the to-be-detected areas are overlapped to form an occupied grid map comprising a plurality of grids, and the grids comprise one or more overlapped areas; and an obstacle detecting device for an aircraft according to any of the embodiments described above.

According to a fifth aspect of the present disclosure, there is provided a computer-storable medium having stored thereon computer program instructions which, when executed by a processor, implement the method for obstacle detection for an aircraft according to any of the embodiments described above.

In the embodiment, the accuracy of detecting the obstacle of the aircraft can be improved, and the efficiency of flying and obstacle avoidance of the aircraft can be further improved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description, serve to explain the principles of the disclosure.

The present disclosure may be more clearly understood from the following detailed description, taken with reference to the accompanying drawings, in which:

FIG. 1 is a flow chart illustrating an obstacle detection method for an aircraft, according to some embodiments of the present disclosure;

FIG. 2 is a projection diagram showing a plurality of radar sensors and an occupancy grid map in a vertical beam direction, according to some embodiments of the present disclosure;

FIG. 3 is a block diagram illustrating an obstacle detection device for an aircraft, according to some embodiments of the present disclosure;

FIG. 4 is a block diagram illustrating an obstacle detection device for an aircraft according to further embodiments of the present disclosure;

FIG. 5 is a block diagram illustrating an obstacle detection system for an aircraft, according to some embodiments of the present disclosure;

FIG. 6 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

Detailed Description

Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

An obstacle detection method for an aircraft according to some embodiments of the present disclosure will be described in detail below with reference to fig. 1 and 2.

Fig. 1 is a flow chart illustrating an obstacle detection method for an aircraft, according to some embodiments of the present disclosure.

Fig. 2 is a projection diagram illustrating a plurality of radar sensors and an occupancy grid map in a vertical beam direction, according to some embodiments of the present disclosure.

As shown in fig. 1, the obstacle detection method for an aircraft includes steps S110 to S130. For example, an obstacle detection method for an aircraft is performed by an obstacle detection device for an aircraft. In some embodiments, the aerial vehicle is a drone.

In step S110, detection results obtained by detecting a plurality of to-be-detected regions located in a specified direction at the current time by a plurality of radar sensors mounted on an aircraft are acquired. A beam emitted by a radar sensor forms an area to be detected. And each detection result represents whether an obstacle exists in the corresponding area to be detected. The multiple regions to be detected are overlapped to form an occupied grid map comprising multiple grids. The plurality of grids includes one or more overlapping regions. In some embodiments, the plurality of grids further comprises a plurality of non-overlapping regions. For example, the number of non-overlapping areas is the same as the number of radar sensors.

In some embodiments, the beam directions of the beams transmitted by the plurality of radar sensors (beam directions at the beam centers) are the same. In other embodiments, the beam directions of the beams emitted by the plurality of radar sensors may also be different or partially different. When there are different beam directions, the adjustment of the relative angle of the beam direction can be performed by those skilled in the art according to the actual situation.

In some embodiments, the plurality of radar sensors is two radar sensors, the designated direction is a flight direction of the aircraft, and the overlap region is one, the overlap region being directly in front of the aircraft and enabling the aircraft to fly through the overlap region. For example, by controlling the maximum width of the overlap region (the overlap region is the cone region, and the maximum width is the diameter of the base circle of the cone region) to be larger than the maximum width of the fuselage of the aircraft by a specified value, it is ensured that the aircraft can fly through the overlap region.

The following describes the arrangement of two radar sensors with reference to fig. 2, taking the two radar sensors as an example.

As shown in fig. 2, two radar sensors S1 and S2 are mounted on the mounting surface of the aircraft. The mounting surface of the aircraft is, for example, a section of the aircraft nose perpendicular to the flight direction.

In some embodiments, twoThe detection distances of the radar sensors S1 and S2 are D1 and D2, respectively. For example, D1 equals D2. The beam directions of the beams transmitted by the radar sensors S1 and S2 (beam directions at the beam centers) are the same. For example, the beam direction is the flight direction. The overlapping configuration of the radar sensor S1 and the radar sensor S2 includes a plurality of grids m_t1、m_t2、m_t3Is (non-standard) occupies the grid map.

The distance between the center of the radar sensor S1 and the center of the radar sensor S2 is L1. For example, the overlap area m may be adjusted by adjusting the position of the center of the radar sensor S1 and/or the position of the center of the radar sensor S2 (i.e., adjusting the distance L1 between the center of the radar sensor S1 and the center of the radar sensor S2)_t2The size of (2). Also for example, the overlap region m may also be adjusted by adjusting the relative angle between the beam directions_t2The size of (2). In some embodiments, the position of the overlap region relative to the center of the aircraft nose (e.g., straight ahead, left-side-ahead, or right-side-ahead) is adjusted by adjusting the distance between the center of the radar sensor S1 or the center of the radar sensor S2 and the center of the aircraft nose.

The beam angle (horizontal detection angle) of the radar sensors S1 and S2 is θ₁M of the overlapping area_t2The detection angle being theta₂，θ₂＝2π-(2π-θ₁) For example, the overlap detection position M is spaced from the mounting surface of the aircraft by a distance ofFor example, the grid m may be judged_t2In the case of an obstacle, the control device of the aircraft may control the aircraft flight obstacle avoidance according to D3.

In step S120, according to the obtained detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located, the forward modeling is used to determine the actual occupation probability that the designated grid is occupied and the actual non-occupation probability that the designated grid is not occupied. For example, the designated grid is an overlap region. Taking FIG. 2 as an example, the designated grid is grid m located directly in front of the aircraft_t2(overlap region). As another example, the designated grid may also be a non-overlapping region. Also by way of example in fig. 2, the designated grid is the grid m located in front of the left side of the aircraft_t1(non-overlapping region).

In some embodiments, the actual occupation probability that the designated grid is occupied is determined by using a forward modeling according to the acquired detection result of the at least one radar sensor corresponding to the to-be-detected region where the designated grid is located and a plurality of preset conditional occupation probabilities of the at least one radar sensor.

The conditional occupancy probability of each of the aforementioned at least one radar sensor comprises a first conditional occupancy probability and a second conditional occupancy probability. The first conditional occupation probability represents the probability that an obstacle exists in the area to be detected where the designated grid is located according to the detection result of the radar sensor under the condition that the designated grid is occupied. The second conditional occupation probability represents the probability that the detection result of the radar sensor is that no obstacle exists in the area to be detected where the designated grid is located under the condition that the designated grid is occupied. The first conditional occupancy probability is greater than the second conditional occupancy probability. The first conditional occupation probability and the second conditional occupation probability are predetermined conditional occupation probabilities based on a detection principle of the radar sensor.

In some embodiments, the first conditional occupancy probability is represented as P (z)_j＝1|m_i1-0.70, and the second conditional probability of occupation is denoted as P (z)_j＝0|m_i1) 0.30, wherein z_jRepresents the detection result of the radar sensor j, m_iRepresenting a grid m_iWhether it is occupied (m)_iWhether an obstacle is present). z is a radical of_j1 denotes that the detection result of the radar sensor is a designated grid m_iIn the area to be detected, there is an obstacle, z_j0 denotes that the detection result of the radar sensor is the designated grid m_iNo obstacle exists in the area to be detected. m is_i1 denotes a grid m_iIs occupied. Taking fig. 2 as an example, j is { S1, S2} and i is { t1, t2, t3 }.

For example, the method includes determining an actual occupation probability that the designated grid is occupied by using a forward modeling according to the acquired detection result of at least one radar sensor corresponding to the to-be-detected region where the designated grid is located and a plurality of preset conditional occupation probabilities of the at least one radar sensor.

First, an occupancy value for a specified occupancy grid is determined. The occupancy value is a ratio of a priori occupancy probability and a priori non-occupancy probability of a preset designated occupancy grid. The prior occupancy probability is less than the prior non-occupancy probability. We first assume that each grid is unoccupied, based on which the a priori probability of occupancy is generally less than the a priori probability of non-occupancy, again because the airspace environment is generally cleaner. For example, the prior occupancy probability is represented as P (m ═ 1) ═ 0.4, and the prior non-occupancy probability is represented as P (m ═ 0) ═ 0.6. Occupancy value ofEach grid shares an occupancy value.

Then, for each of the at least one radar sensor, a conditional occupancy probability corresponding to the acquired detection result is selected from the conditional occupancy probabilities of the radar sensor according to the acquired detection result.

In some embodiments, for each of the aforementioned at least one radar sensor, in the case that the obtained detection result is that an obstacle exists, the selected corresponding conditional occupancy probability is the first conditional occupancy probability; and under the condition that the obtained detection result is that no obstacle exists, the selected corresponding conditional occupation probability is the second conditional occupation probability.

Taking FIG. 2 as an example, the designated grid is grid m_t2In the case of (2), grid m_t2The region to be detected is a grid m_t1And a grid m_t2The radar sensors corresponding to the regions are S1 and S2. For example, if the detection results of the radar sensors S1 and S2 at the current time are 1 (obstacle present) and 0 (obstacle absent), respectively, the conditional occupancy probability corresponding to the detection result 1 of the radar sensor S1 is the first conditional occupancy probability P (z < z >)_S1＝1|m_t21 — 0.70, the conditional occupancy probability corresponding to the detection result 0 of the radar sensor S2 is the second conditional occupancy probability P (z)_S2＝0|m_t21) 0.30. The sum of the first conditional occupation probability and the second conditional occupation probability is 1.

And finally, determining the actual occupation probability of the appointed grid according to the conditional occupation probability and the occupation value of the selected at least one radar sensor.

In some embodiments, in the case that there is only one of the at least one radar sensor, the product of the conditional occupancy probability of the selected radar sensor and the occupancy value is determined as the actual occupancy probability that the designated grid is occupied.

In other embodiments, in the case that there are a plurality of the at least one radar sensor, the actual occupancy probability that the designated grid is occupied is determined as the product of the sum of the conditional occupancy probabilities of the selected radar sensors and the occupancy value.

Taking the designated grid as the grid m of FIG. 2_t2The detection results of the radar sensors S1 and S2 at the current time are 1 (presence of an obstacle) and 0 (absence of an obstacle), respectively, for example, grid m_t2Actual probability of being occupied

Similarly, the designated grid is the grid m of FIG. 2_t1The detection results of the radar sensors S1 and S2 at the current time are 1 (presence of an obstacle) and 0 (absence of an obstacle), respectively, for example, grid m_t1Actual probability of being occupied Likewise, grid m_t3Actual probability of being occupied

In some embodiments, the actual non-occupation probability that each grid is not occupied is determined by using a forward modeling according to the acquired detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located and a plurality of conditional non-occupation probabilities of the at least one radar sensor.

The conditional non-occupancy probability for each of the aforementioned at least one radar sensor comprises a first conditional non-occupancy probability and a second conditional non-occupancy probability. The first conditional non-occupation probability represents the probability that the detection result of the radar sensor is that no obstacle exists in the area to be detected where the designated grid is located under the condition that the designated grid is not occupied. The second conditional non-occupation probability represents the probability that the detection result of the radar sensor is that an obstacle exists in the area to be detected where the designated grid is located under the condition that the designated grid is not occupied. The first conditional non-occupancy probability is greater than the second conditional non-occupancy probability. The first conditional non-occupancy probability is greater than the first conditional occupancy probability. The second conditional non-occupancy probability is less than the second conditional occupancy probability. The first conditional non-occupancy probability and the second conditional non-occupancy probability are conditional non-occupancy probabilities preset based on a detection principle of the radar sensor.

In some embodiments, the first conditional non-occupancy probability is denoted as P (z)_j＝1|m_i0) 0.80, and the second conditional non-occupation probability is denoted as P (z)_j＝0|m_i0) 0.20, wherein z_jRepresents the detection result of the radar sensor j, m_iRepresenting a grid m_iWhether it is occupied (m)_iWhether an obstacle is present). z is a radical of_j1 denotes that the detection result of the radar sensor is a designated grid m_iIn the area to be detected, there is an obstacle, z_j0 denotes that the detection result of the radar sensor is the designated grid m_iNo obstacle exists in the area to be detected. m is_i1 denotes a grid m_iIs occupied. Taking fig. 2 as an example, j is { S1, S2} and i is { t1, t2, t3 }.

For example, the actual non-occupation probability that the designated grid is not occupied is determined by utilizing a forward modeling according to the acquired detection result of at least one radar sensor corresponding to the region to be detected where the designated grid is located.

First, for each of the aforementioned at least one radar sensor, a conditional non-occupancy probability corresponding to the acquired detection result is selected from the conditional non-occupancy probabilities of the radar sensor according to the acquired detection result.

In some embodiments, for each of the aforementioned at least one radar sensor, in the case that the obtained detection result is that no obstacle exists, the selected corresponding conditional non-occupancy probability is the first conditional non-occupancy probability; and in the case that the acquired detection result is that the obstacle exists, the selected corresponding conditional non-occupation probability is the second conditional non-occupation probability.

Taking FIG. 2 as an example, the designated grid is grid m_t2In the case of (2), grid m_t2The region to be detected is a grid m_t1And a grid m_t2The radar sensors corresponding to the regions are S1 and S2. For example, if the detection results of the radar sensors S1 and S2 at the current time are 1 (obstacle present) and 0 (obstacle absent), respectively, the conditional non-occupancy probability corresponding to the detection result 1 of the radar sensor S1 is the first conditional non-occupancy probability P (z_S1＝1|m_t20) to 0.80, the conditional occupancy probability corresponding to the detection result 0 of the radar sensor S2 is the second conditional non-occupancy probability P (z)_S2＝0|m_t20) 0.20. The sum of the first conditional non-occupancy probability and the second conditional non-occupancy probability is 1.

Then, an actual non-occupancy probability is determined that the designated grid is unoccupied, based on the selected conditional non-occupancy probability of the aforementioned at least one radar sensor.

In some embodiments, in the case of only one of the aforementioned at least one radar sensor, the selected conditional non-occupancy probability of that radar sensor is determined as the actual non-occupancy probability that the designated grid is unoccupied.

In other embodiments, in the case that there are a plurality of the aforementioned at least one radar sensor, the sum of the conditional non-occupancy probabilities of the selected individual radar sensors is determined as the actual non-occupancy probability that the designated grid is not occupied.

Taking the designated grid as the grid m of FIG. 2_t2The detection results of the radar sensors S1 and S2 at the current time are 1 (presence of an obstacle) and 0 (absence of an obstacle), respectively, for example, grid m_t2Actual unoccupied probability P (m) of unoccupied_t2＝0)＝P(z_S1＝1|m_t2＝0)+P(z_S2＝0|m_t2＝0)＝0.2+0.8＝1。

Similarly, the designated grid is the grid m of FIG. 2_t1The detection results of the radar sensors S1 and S2 at the current time are 1 (presence of an obstacle) and 0 (absence of an obstacle), respectively, for example, grid m_t1Actual unoccupied probability P (m) of unoccupied_t1＝0)＝P(z_S1＝1|m_t10) 0.2. Likewise, grid m_t3Actual unoccupied probability P (m) of unoccupied_t3＝0)＝P(z_S2＝0|m_t3＝0)＝0.8。

In the process, the principle of calculating the actual occupation probability and the actual non-occupation probability of each grid is based on the forward modeling principle of the occupied grids. In the forward model principle of the occupied grid, the map posterior can be decomposed into a map prior and a measurement likelihood, and the map posterior can be approximated to the actual occupation probability or the actual non-occupation probability of the application, so that the actual occupation probability or the actual non-occupation probability of the application can be decomposed into a conditional occupation probability or a conditional non-occupation probability (measurement likelihood) based on a measurement value and an occupation value (map prior) obtained by the prior occupation probability and the prior non-occupation probability. The forward model can maximize the posterior probability and improve the accuracy of calculation.

In step S130, for the designated grid, it is determined whether there is an obstacle in the designated grid according to the corresponding actual occupancy probability and actual non-occupancy probability.

In some embodiments, the magnitude relationship of the logarithm of the actual probability of occupation and the logarithm of the actual probability of non-occupation is compared; and determining whether the specified grid has the obstacle according to the comparison result.

For example, the base number of the logarithm is greater than 1, and in the case that the logarithm of the actual occupation probability is greater than the logarithm of the actual non-occupation probability as a result of the comparison, it is determined that an obstacle exists in the designated grid; and determining that the specified grid has no obstacle when the logarithm of the actual occupation probability is smaller than the logarithm of the actual non-occupation probability as a result of the comparison.

For example, the base of the logarithm is 10. With the grid m of FIG. 2_t2For the purpose of example only, log₁₀P(m_t2＝0)＝log₁₀1 is 0. By comparison, log₁₀P(m_t21) less than log₁₀P(m_t20). Thus, grid m_t2No obstacles are present. In some embodiments, the logarithm of the actual occupancy probability or the logarithm of the actual non-occupancy probability may be directly calculated by first taking the logarithm of the occupancy value, the conditional occupancy probability, or the conditional non-occupancy probability. The logarithm taking is carried out according to the forward modeling principle of occupying grids, and is set for convenient calculation.

In some embodiments, the obstacle detection method for an aircraft further comprises: after determining whether the specified grid has the obstacle, establishing a corresponding relation between detection results of a plurality of radar sensors and a result of determining whether the specified grid has the obstacle; and under the condition of obtaining detection results obtained by detecting a plurality of areas to be detected by a plurality of radar sensors at subsequent time, determining whether the specified grid has an obstacle or not according to the established corresponding relation. By establishing the known corresponding relation, the efficiency of obstacle detection can be improved, more time is reserved for the flight obstacle avoidance of the aircraft, and therefore the efficiency of the flight obstacle avoidance of the aircraft is further improved. It should be understood by those skilled in the art that in the case where a certain correspondence cannot be found, the process from the foregoing step S110 to step S130 should be adopted to determine whether an obstacle exists.

Taking fig. 2 as an example, after multiple detections, a table of correspondence between detection results and whether obstacles exist in the grid may be established as shown in table 1.

TABLE 1 table of correspondence between detection results and whether obstacles exist in the grid

	zs1＝1，zs2＝0	zs1＝1，zs2＝1	zs1＝0，zs2＝0	zs1＝0，zs2＝1
					mt1	Presence of obstacles	Presence of obstacles	Absence of obstacles	Absence of obstacles
mt2	Absence of obstacles	Presence of obstacles	Absence of obstacles	Absence of obstacles
					mt3	Absence of obstacles	Presence of obstacles	Absence of obstacles	Presence of obstacles

In the above embodiment, the plurality of radar sensors are arranged on the aircraft, the regions to be detected corresponding to the plurality of radar sensors are overlapped to form a (nonstandard) occupation grid map, the plurality of regions to be detected are subdivided and divided, and then the specified grid (the region concerned by the aircraft in flight) after the subdivision is determined whether an obstacle exists by combining the nonstandard occupation grid map with the forward modeling model, so that the accuracy of obstacle detection of the aircraft can be improved, a more refined basis is provided for the aircraft in flight obstacle avoidance, and the efficiency of the aircraft in flight obstacle avoidance can be improved.

Fig. 3 is a block diagram illustrating an obstacle detection device for an aircraft according to some embodiments of the present disclosure.

As shown in fig. 3, the obstacle detecting apparatus 31 for an aircraft includes an acquisition module 311, a first determination module 312, and a second determination module 313.

The obtaining module 311 is configured to obtain detection results obtained by detecting a plurality of to-be-detected regions located in a specified direction at the current time by a plurality of radar sensors installed on the aircraft, for example, step S110 shown in fig. 1 is performed. A beam emitted by a radar sensor forms a region to be detected, each detection result represents whether an obstacle exists in the corresponding region to be detected, a plurality of regions to be detected are overlapped to form an occupied grid map comprising a plurality of grids, and the plurality of grids comprise one or more overlapped regions.

The first determining module 312 is configured to determine, according to the obtained detection result of the at least one radar sensor corresponding to the to-be-detected region where the designated grid is located, an actual occupation probability that the designated grid is occupied and an actual non-occupation probability that the designated grid is not occupied by using a forward model, for example, step S120 shown in fig. 1 is performed.

The second determination module 313 is configured to determine, for a specified grid, whether there is an obstacle in the specified grid according to the corresponding actual occupancy probability and actual non-occupancy probability, for example, to perform step S130 as shown in fig. 1.

FIG. 4 is a block diagram illustrating an obstacle detection device for an aircraft according to further embodiments of the present disclosure.

As shown in fig. 4, the obstacle detecting device 41 for an aircraft includes a memory 411; and a processor 412 coupled to the memory 411. The memory 411 is used to store instructions to perform a corresponding embodiment of the method of obstacle detection for an aircraft. The processor 412 is configured to execute the obstacle detection method for an aircraft in any of the embodiments of the present disclosure based on instructions stored in the memory 411.

FIG. 5 is a block diagram illustrating an obstacle detection system for an aircraft, according to some embodiments of the present disclosure.

As shown in fig. 5, the obstacle detecting system 5 for an aircraft includes a plurality of radar sensors 50A, 50B, etc., and an obstacle detecting device 51 for an aircraft.

A plurality of radar sensors 50A, 50B, etc. are mounted on the aircraft. The plurality of radar sensors 50A, 50B, etc. are configured to detect a plurality of areas to be detected located in a specified direction at the present time to obtain detection results. A beam emitted by one radar sensor forms a region to be detected, and each detection result represents whether an obstacle exists in the corresponding region to be detected. The multiple regions to be detected are overlapped to form an occupied grid map comprising multiple grids. The plurality of grids includes one or more overlapping regions.

The obstacle detecting device 51 for an aircraft is, for example, the same as or similar in structure and function to the obstacle detecting device 31 for an aircraft in fig. 3 and the obstacle detecting device 41 for an aircraft in fig. 4.

FIG. 6 is a block diagram illustrating a computer system for implementing some embodiments of the present disclosure.

As shown in FIG. 6, computer system 60 may take the form of a general purpose computing device. Computer system 60 includes a memory 610, a processor 620, and a bus 600 that connects the various system components.

The memory 610 may include, for example, system memory, non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs. The system memory may include volatile storage media such as Random Access Memory (RAM) and/or cache memory. The non-volatile storage medium stores, for instance, instructions to perform corresponding embodiments of at least one of the obstacle detection methods for the aircraft. Non-volatile storage media include, but are not limited to, magnetic disk storage, optical storage, flash memory, and the like.

The processor 620 may be implemented as discrete hardware components, such as a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gates or transistors, or the like. Accordingly, each of the modules, such as the judging module and the determining module, may be implemented by a Central Processing Unit (CPU) executing instructions in a memory for performing the corresponding step, or may be implemented by a dedicated circuit for performing the corresponding step.

Bus 600 may use any of a variety of bus architectures. For example, bus structures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.

Computer system 60 may also include input-output interface 630, network interface 640, storage interface 650, and the like. These interfaces 630, 640, 650 and the memory 610 and the processor 620 may be connected by a bus 600. The input/output interface 630 may provide a connection interface for input/output devices such as a display, a mouse, and a keyboard. The network interface 640 provides a connection interface for various networking devices. The storage interface 650 provides a connection interface for external storage devices such as a floppy disk, a usb disk, and an SD card.

Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable apparatus to produce a machine, such that the execution of the instructions by the processor results in an apparatus that implements the functions specified in the flowchart and/or block diagram block or blocks.

These computer-readable program instructions may also be stored in a computer-readable memory that can direct a computer to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and/or block diagram block or blocks.

The present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.

By the obstacle detection method, the obstacle detection device and the obstacle detection system for the aircraft and the computer-storable medium in the embodiments, the accuracy of obstacle detection of the aircraft can be improved, and the efficiency of flight obstacle avoidance of the aircraft can be further improved.

Thus far, the obstacle detection method, apparatus and system, computer-readable storage medium for an aircraft according to the present disclosure have been described in detail. Some details that are well known in the art have not been described in order to avoid obscuring the concepts of the present disclosure. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.