WO2021217341A1

WO2021217341A1 - Obstacle avoidance method, moveable platform, control device, and storage medium

Info

Publication number: WO2021217341A1
Application number: PCT/CN2020/087239
Authority: WO
Inventors: 高迪; 王俊喜; 王石荣
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2020-04-27
Filing date: 2020-04-27
Publication date: 2021-11-04
Also published as: CN112753000A

Abstract

Provided in the embodiments of the present application are an obstacle avoidance method, a moveable platform, a control device, and a storage medium. The method comprises: a moveable platform controlling the rotation of a radar configured on the moveable platform, and the radar being able to collect point cloud data during the rotation; at this time, a control device acquiring omnidirectional point cloud data corresponding to the operating environment of the moveable platform; then, according to the omnidirectional point cloud data, determining whether there is an obstacle in the operating environment of the moveable platform; and if there is an obstacle, controlling the moveable platform to avoid the obstacle. It can be seen that in the obstacle avoidance method, a movable platform can realize an omnidirectional obstacle avoidance function thereof using point cloud data collected by a radar, thereby avoiding a poor obstacle avoidance effect caused by mutual interference between multiple obstacle avoidance sensors while ensuring the obstacle avoidance effect, and greatly reducing the risk of damaging the movable platform.

Description

Obstacle avoidance method, movable platform, control equipment and storage medium

Technical field

The invention relates to the field of obstacle avoidance, in particular to an obstacle avoidance method, a movable platform, a control device and a storage medium.

Background technique

In recent years, mobile platforms such as autonomous vehicles and unmanned aerial vehicles have been widely used in industry and daily life. In order to ensure the normal operation of the movable platform, it often needs to have an omnidirectional obstacle avoidance function.

In the prior art, the movable platform usually uses multiple obstacle avoidance sensors arranged at different positions to obtain a full range of environmental information, so as to realize obstacle avoidance. Among them, the obstacle avoidance sensor can be a vision sensor, an ultrasonic sensor, a radar sensor, and so on. However, the use of multiple obstacle avoidance sensors often fails to make the movable platform have a better obstacle avoidance effect, which makes the movable platform have a greater risk of damage during operation.

Summary of the invention

The invention provides an obstacle avoidance method, a movable platform, a control device and a storage medium, which are used to ensure the obstacle avoidance effect.

The first aspect of the present invention is to provide an obstacle avoidance method, which includes:

Controlling the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment where the movable platform is located;

Determine whether there are obstacles in the operating environment according to the point cloud data;

If there are obstacles in the operating environment, the movable platform is controlled to avoid obstacles.

The second aspect of the present invention is to provide a movable platform, which at least includes: a body, a power system, a radar, and a control device;

The power system is arranged on the body and used to provide power for the movable platform;

The radar is arranged on the body and is used to collect point cloud data;

The control device includes a memory and a processor;

The memory is used to store a computer program;

The processor is configured to run a computer program stored in the memory to realize:

Controlling the rotation of the radar configured on the movable platform to obtain the omnidirectional point cloud data in the operating environment where the movable platform is located;

The third aspect of the present invention is to provide a control device including:

Memory, used to store computer programs;

The fourth aspect of the present invention is to provide a computer-readable storage medium, the storage medium is a computer-readable storage medium, the computer-readable storage medium stores program instructions, and the program instructions are used in the first aspect. The obstacle avoidance method described.

The invention provides an obstacle avoidance method, a movable platform, a control device and a storage medium. The movable platform controls the rotation of the radar configured by itself, and the radar can collect point cloud data during the rotation. At this point, the control device can obtain all-round point cloud data in the operating environment of the mobile platform. Then, according to the omnidirectional point cloud data, it is determined whether there are obstacles in the operating environment of the movable platform. If there are obstacles, control the movable platform to avoid obstacles.

It can be seen that in the above obstacle avoidance method, the mobile platform can use the point cloud data collected by a radar to achieve its own omni Interference with each other, resulting in poor obstacle avoidance, greatly reducing the risk of damage to the movable platform.

Description of the drawings

The drawings described here are used to provide a further understanding of the application and constitute a part of the application. The exemplary embodiments and descriptions of the application are used to explain the application, and do not constitute an improper limitation of the application. In the attached picture:

Figure 1 is a schematic structural diagram of an unmanned aerial vehicle provided by an embodiment of the present invention;

FIG. 2 is a flowchart of an obstacle avoidance method provided by an embodiment of the present invention;

FIG. 3a is a schematic diagram of an orientation division according to an embodiment of the present invention;

Figure 3b is a schematic diagram of another orientation division provided by an embodiment of the present invention;

4 is a flowchart of another obstacle avoidance method provided by an embodiment of the present invention;

FIG. 5 is a flowchart of another obstacle avoidance method provided by an embodiment of the present invention;

Figure 6a is a schematic diagram of a first trajectory of the movable platform;

Figure 6b is a schematic diagram of a second trajectory of the movable platform;

Figure 6c is a schematic diagram of another first trajectory of the movable platform;

FIG. 7 is a schematic structural diagram of a control device provided by an embodiment of the present invention;

FIG. 8 is a schematic structural diagram of a movable platform provided by an embodiment of the present invention;

Fig. 9 is a schematic structural diagram of a control device provided by an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

In practical applications, the movable platform mentioned in the present invention may be an unmanned aerial vehicle, an autonomous vehicle, an unmanned boat, or an intelligent robot with mobile capabilities, and so on. For the normal movement of the equipment, the obstacle avoidance function is one of the important functions that it needs to have. When the movable platform is specifically an unmanned aerial vehicle, it can be applied to various fields such as electric power, agriculture, film and television entertainment. The schematic diagram of the structure of the unmanned aerial vehicle can be shown in Figure 1.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following embodiments, the movable platform is an unmanned aerial vehicle as an example for description. And if there is no conflict between the embodiments, the following embodiments and the features in the embodiments can be combined with each other.

FIG. 2 is a schematic flowchart of an obstacle avoidance method provided by an embodiment of the present invention. The main body of execution of this obstacle avoidance method is the control device. It can be understood that the control device can be implemented as software or a combination of software and hardware. The control device can make the movable platform realize obstacle avoidance by executing the obstacle avoidance method. The control device in this embodiment and the following embodiments may specifically be a control device, and the control device may specifically be a controller configured on a movable platform, or a remote server. Specifically, the method may include:

S101: Control the rotation of the radar configured on the movable platform to obtain a full range of point cloud data in the operating environment where the movable platform is located.

The movable platform is equipped with a radar. After the movable platform is turned on, the control device can control the radar to turn on to collect point cloud data through the radar. Optionally, the radar configured on the movable platform may be a millimeter wave radar, and the rotation axis of the radar may be parallel to the yaw axis of the movable platform.

Optionally, one of the most common methods: the radar can rotate 360 degrees around the rotation axis, so as to collect all-round point cloud data in the operating environment of the movable platform during continuous rotation.

S102: Determine whether there is an obstacle in the operating environment according to the point cloud data.

After the control device obtains a full range of point cloud data, it can optionally perform clustering processing on the point cloud data through its own preset clustering algorithm, and determine the current operation of the movable platform according to the clustering result. Whether there are obstacles in the environment. Wherein, the clustering result may include at least one cluster. If the number of point cloud data contained in the target cluster is greater than or equal to the preset number, it can be determined that there is an obstacle in the operating environment, that is, the flight environment, and the target cluster may be at least one cluster. Any one of, this target cluster corresponds to an obstacle, that is, the point cloud data in the target cluster is used to describe the obstacle. If in at least one cluster, the number of point cloud data contained in each cluster is less than the preset number, it can be determined that the current operating environment of the mobile platform does not contain obstacles.

Optionally, the clustering algorithm used above may be a density-based clustering (Density-Based Spatial Clustering of Applications with Noise, DBSCAN for short) algorithm with noise. Of course, the preset clustering algorithm may also be any mature algorithm in the prior art, such as the k-means clustering algorithm, that is, the K-means algorithm and so on.

S103: If there are obstacles in the operating environment, control the movable platform to avoid obstacles.

If it is determined based on the clustering result that there is an obstacle in the operating environment, the control device may further determine the positional relationship between the obstacle and the movable platform based on the point cloud data in the target cluster corresponding to the obstacle. Then plan the running path for the movable platform according to the path planning algorithm configured by the control device itself, so that the movable platform runs according to the planned path and avoids obstacles.

In this embodiment, the movable platform controls the rotation of the radar configured by itself, and the radar can collect point cloud data during the rotation. At this point, the control device can obtain all-round point cloud data in the operating environment of the mobile platform. Then, according to the omnidirectional point cloud data, it is determined whether there are obstacles in the operating environment of the movable platform. If there are obstacles, control the movable platform to avoid obstacles. It can be seen that in the above obstacle avoidance method, the mobile platform can use the point cloud data collected by a radar to achieve its own omni Interference with each other, resulting in poor obstacle avoidance, greatly reducing the risk of damage to the movable platform.

It should be noted that the foregoing embodiment illustrates that the radar can continuously rotate 360 degrees, so as to continuously collect corresponding point cloud data in the 360-degree operating environment of the movable platform. The control device can only start to determine whether there are obstacles in the operating environment after acquiring the 360-degree point cloud data.

However, in order to improve the real-time performance of obstacle determination, optionally, the radar can rotate intermittently around the rotation axis, and the radar stays for a preset time when it rotates to a preset angle, and collects point cloud data within the preset time. And based on the point cloud data collected within the preset time to determine obstacles. For example, the radar can sequentially collect the point cloud data corresponding to at least one azimuth of the movable platform through intermittent rotation, and determine the obstacle after obtaining the point cloud data of at least one azimuth.

Wherein, the aforementioned at least one orientation may be at least one of front, back, left, and right, as shown in Fig. 3a. Each azimuth corresponds to an area of 90 degrees, and the corresponding relationship between the preset angle and the azimuth can be considered as front-45 degrees, right-135 degrees, rear--225 degrees, and left-275 degrees. Of course, the division of the orientation or the preset angle is not limited to the above-mentioned "front, rear, left, and right" division method, and may also have a smaller division granularity, as shown in FIG. 3b.

It is easy to understand that the obstacle can be located in any position of the movable platform, and the determination of whether there is an obstacle is based on the point cloud data. Therefore, in order to ensure the accuracy and effectiveness of the point cloud data, FIG. 4 is a schematic flowchart of another obstacle avoidance method provided by an embodiment of the present invention. As shown in Figure 4, the obstacle avoidance method may further include the following steps:

S201: Control the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment where the movable platform is located.

The execution process of the foregoing step 201 is similar to the corresponding steps of the foregoing embodiment, and reference may be made to the related description in the embodiment shown in FIG. 2, which will not be repeated here.

S202: Filter the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data.

In practical applications, some obstacles far away from the movable platform can be considered as obstacles outside the safe movement range of the movable platform. The point cloud data used to describe these invalid obstacles can also be considered as invalid data, and the invalid data can be filtered out before the clustering process. The point cloud data remaining after filtering is the effective point cloud data used for obstacle determination, that is, the target point cloud data.

Optionally, taking an unmanned aerial vehicle as an example, the safe movement range of the unmanned aerial vehicle may be a space range formed by 0.5 meters below the airframe and 2 meters above the airframe.

S203: Perform clustering processing on the target point cloud data to obtain at least one cluster.

S204: Determine whether the number of point cloud data contained in the target cluster is greater than or equal to a preset number, the target cluster is any one of at least one cluster, if it is greater, step 205 to step 207 are executed, otherwise, step 208 is executed.

S205: Determine that there is an obstacle in the operating environment.

S206: Determine the size information of the obstacle according to the point cloud data in the target cluster.

S207: Control the movable platform to avoid the obstacle according to the size information of the obstacle.

S208: It is determined that there are no obstacles in the operating environment, and the movable platform moves in the current direction of movement.

Then, the control device can cluster the target point cloud data. Wherein, the clustering algorithm may be any one mentioned in the embodiment shown in FIG. 2, and the clustering result may include at least one cluster. For any one of clusters A, that is, the target cluster, if the number of point cloud data contained in cluster A is greater than or equal to the preset number, it is determined that there are obstacles in the operating environment of the movable platform, and this cluster A is used for Describe an obstacle. After performing the above judgment on each cluster, it can be determined that there are several obstacles in the movement environment of the movable platform.

Then, optionally, the point cloud data may be specifically represented as three-dimensional coordinates. Therefore, the size information of the obstacle, such as the length, width, and height of the obstacle, may also be determined according to the three-dimensional coordinates in the target cluster.

Then, the control device can also plan a path for the movable platform according to the path planning algorithm configured by itself, combined with the size information of the obstacle, so that the movable platform can avoid the obstacle with the shortest running distance.

In at least one cluster included in the clustering result, if the amount of point cloud data included in each cluster is less than the preset amount, it is determined that there is no obstacle in the operating environment. At this point, the movable platform can continue to move in the current posture.

It should be noted that steps 203 to 204 in this embodiment are not described in detail, and reference may also be made to related descriptions in the embodiment shown in FIG. 2, which will not be repeated here.

In this embodiment, by filtering the point cloud data, the target point cloud data that is closely related to the safe operation of the mobile platform can be retained. The control device then judges obstacles according to the retained target point cloud data, so as to realize obstacle avoidance of the movable platform. Compared with all the point cloud data, the amount of target point cloud data is greatly reduced, which can improve the efficiency of the control equipment to determine obstacles, ensure the accuracy and timeliness of obstacle avoidance, and avoid damage to the movable platform.

Based on the description of the foregoing embodiments, it can be known that the point cloud data collected by the radar can reflect the position of the obstacle. In order to enable the point cloud data to accurately reflect the position of the obstacle, so that it will not be disturbed by the movement posture of the movable platform, therefore, FIG. 5 is a schematic flowchart of another obstacle avoidance method provided by an embodiment of the present invention. As shown in Figure 5, the obstacle avoidance method may further include the following steps:

S301: Control the rotation of the radar configured on the movable platform to obtain all-round point cloud data in the operating environment where the movable platform is located.

The execution process of the foregoing step 301 is similar to the corresponding steps of the foregoing embodiment, and reference may be made to the related description in the embodiment shown in FIG. 2, which will not be repeated here.

S302: Convert the three-dimensional coordinates of the point cloud data from the spherical coordinate system to the body coordinate system of the movable platform.

Perform coordinate conversion on the acquired point cloud data. The point cloud data can be expressed as three-dimensional coordinates. Optionally, the three-dimensional coordinates of the point cloud data are specifically spherical coordinates. In this case, you need to convert the spherical coordinates to rectangular coordinates first. Make the three-dimensional coordinates of the converted point cloud data correspond to the rectangular coordinate system. Taking the movable platform as an unmanned aerial vehicle as an example, the rectangular coordinate system may be the airframe coordinate system. Among them, the body coordinate system can be defined as: the center of mass of the unmanned aerial vehicle is the origin of the coordinate, the direction of the X axis is located in the reference plane of the unmanned aerial vehicle, parallel to the fuselage axis and pointing to the front of the unmanned aerial vehicle, and the Y axis is perpendicular to the unmanned aerial vehicle The reference plane points to the right of the UAV, and the Z axis is perpendicular to the XOY plane in the reference plane and points above the UAV.

S303: Convert the three-dimensional coordinates from the body coordinate system to the inertial coordinate system to determine whether there is an obstacle in the operating environment according to the converted point cloud data, if there is an obstacle, perform step 304; otherwise, perform step 306.

Further optionally, the three-dimensional coordinates of the point cloud data can be converted from the body coordinate system to the inertial coordinate system. Among them, the inertial coordinate system is the northeast coordinate system. Through the above coordinate conversion, the influence of the flight attitude of the unmanned aerial vehicle on the accuracy of the point cloud data can be offset.

For the converted point cloud data, the control device can also use a clustering algorithm to determine whether there are obstacles in the operating environment of the movable platform. For the specific process, please refer to the relevant description in the embodiment shown in Figure 1 or Figure 4, here No longer.

In addition, the point cloud data collected by the radar may have noise, and the interference of the noise will further affect the accuracy of the obstacle determination result. Therefore, the judgment result of the obstacle can also be verified.

In S304, according to the point cloud data and the movement speed of the movable platform, check whether there is an obstacle in the operating environment. If the check result shows that there is an obstacle, perform step 305; otherwise, perform step 306.

S305: Control the movable platform to avoid obstacles.

S306: Control the movable platform to move along the current direction of movement.

If the verification result shows that there are obstacles in the operating environment of the movable platform, the control movable platform will avoid obstacles according to the path planning algorithm. If the verification result shows that there is no obstacle, it can be considered that the control device has misdetected the obstacle. At this time, the control device will continue to control the movable platform to continue to move in the current direction in the current posture.

For the verification process of obstacle judgment results, the following optional methods can be used:

The radar on the movable platform can rotate around an axis to collect point cloud data. It can be considered that the time required for each rotation of the radar by a preset angle is a collection period, and the preset angle can be, for example, 360 degrees, 180 degrees, and so on. When the movable platform is in the operating state I at time T1, the control device can obtain the point cloud data collected by the radar from time T0 to time T1. Among them, time T0 to time T1 corresponds to the first number of acquisition cycles of the radar.

At this time, the control device can determine whether there is an obstacle in the operating environment of the movable platform according to the point cloud data collected in each collection period, and determine the distance between the obstacle and the movable platform. Then the control device can obtain the first number of distance values, which can be collectively referred to as the first distance value. From this first distance value, the first trajectory of the movable platform from time T0 to time T1 is generated, and this first trajectory can be regarded as a true trajectory.

Optionally, before determining the distance between the movable platform and the obstacle, as described in the embodiment shown in FIG. The filtered point cloud data determines the distance between the obstacle and the movable platform.

When the movable platform is in operating state II at time T2, the control device can obtain the point cloud data collected by the radar from time T0 to time T2. Among them, time T0 to time T2 corresponds to the second number of acquisition cycles of the radar, and time T2 is earlier than time T1, that is, the second number is less than the first number.

At this time, the control device can determine whether there are obstacles in the operating environment of the movable platform based on the point cloud data collected in each collection period, and determine the distance between the obstacle and the movable platform, and the control device can obtain The second number of distance values.

Then, the difference between the first number and the second number is M, then the control device can predict the distance between the movable platform and the obstacle in this M collection period according to the operating speed of the movable platform at time T2. M distance values, and the M distance values and the aforementioned second number of distance values form the first number of distances, and the distance values of these two parts can be collectively referred to as the second distance value. From this second distance value, the second trajectory of the movable platform from time T0 to time T1 is generated. Although both the first track and the second track contain the first number of distance values, M distance values in the second track are predicted values. Therefore, this second track can be regarded as a predicted track.

Also optionally, before determining the distance between the movable platform and the obstacle, as described in the embodiment shown in FIG. 4, the first point cloud data collected in the second number of collection periods can also be filtered. And determine the distance between the obstacle and the movable platform according to the filtered point cloud data.

[Corrected according to Rule 91 04.03.2021]
Furthermore, after the two tracks are obtained, the similarity between the two can be determined. If the similarity meets the preset range, it is determined that there are obstacles in the operating environment of the movable platform. Otherwise, it is considered that there are no obstacles in the operating environment. At this time, the control device can control the movable platform to continue to move in the current motion direction in the current posture. And because there is no need to plan the obstacle avoidance path based on the trajectory at this time. Therefore, the control device can also delete the above-mentioned first track and second track.

For the above verification process, specific examples can also be used to illustrate:

Assume that the moving speed of the movable platform is 1m/s, the radar acquisition period is 1 second, and the first number is 5 and the second number is 3.

Based on the above assumptions, in one case, the control device can determine the distances between the movable platform and the obstacle in front as 10 meters, 9 meters, and 8 meters based on the point cloud data collected by the radar in the 1st to 5th seconds respectively. Meters, 7 meters, 6 meters. At this time, the control device can draw the first track based on the above five actual distance values, as shown in Fig. 6a.

The control device can also respectively determine the distances between the movable platform and the obstacle in front as 10 meters, 9 meters, and 8 meters based on the point cloud data collected by the radar in the first to third seconds. At the same time, the control device can also predict that the distance between the movable platform and the obstacle is 7 meters and 6 meters in the 4th and 5th seconds according to the operating speed of the movable platform. At this time, the second trajectory drawn by the control device based on the above three actual distance values and the two predicted distance values can be as shown in Figure 6b. At this time, the first track and the second track are completely coincident, the actual distance between the movable platform and the obstacle is the same as the predicted distance, and the similarity of the two tracks meets the preset range, indicating the obstacle in front of the movable platform Things are real.

In another case, the control device determines that the distance between the movable platform and the obstacle in front is 10 meters, 9 meters, 8 meters, and 6 meters based on the point cloud data sequentially collected from the first second to the fifth second. 2 meters. From these 5 actual distance values, that is, the first distance value, the first track can be obtained as shown in Fig. 6c.

The control device can also determine the distance between the movable platform and the obstacle in front as 10 meters, 9 meters, and 8 meters based on the point cloud data collected by the radar from the first second to the third second respectively. At the same time, the control device can predict the distance between the movable platform and the obstacle to be 6 meters and 5 meters in the 4th and 5th seconds according to the operating speed of the movable platform. The second trajectory that can be drawn from the three actual distance values and the two predicted distance values can be as shown in Figure 6b. At this time, the similarity between the first trajectory and the second trajectory does not meet the preset range, indicating that the movable platform has misjudged the obstacle.

In this embodiment, the control device can determine whether there are obstacles in the operating environment of the movable platform. Because the judgment result will directly affect the running posture of the movable platform. Therefore, in order to ensure the normal operation of the movable platform, the judgment result of the obstacle can also be verified, and the movable platform can be determined according to the verification result to further adjust the operating posture of the movable platform, so that the movable platform will not appear due to obstacles. Operational errors or even damage caused by incorrect determination of objects.

FIG. 7 is a schematic structural diagram of a control device provided by an embodiment of the present invention; referring to FIG. 7, this embodiment provides a control device that can execute the above obstacle avoidance method; specifically, the control device include:

The control module 11 is used to control the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment where the movable platform is located.

The determining module 12 is configured to determine whether there is an obstacle in the operating environment according to the point cloud data.

The control module 11 is also used to control the movable platform to avoid obstacles if there are obstacles in the operating environment.

The device shown in FIG. 7 can also execute the method of the embodiment shown in FIG. 1 to FIG. 6c. For parts that are not described in detail in this embodiment, please refer to the related description of the embodiment shown in FIG. 1 to FIG. 6c. For the implementation process and technical effects of this technical solution, please refer to the description in the embodiment shown in FIG. 1 to FIG. 6c, which will not be repeated here.

FIG. 8 is a schematic structural diagram of a movable platform provided by an embodiment of the present invention; referring to FIG. 8, an embodiment of the present invention provides a movable platform, and the movable platform is at least one of the following: Aircraft, self-driving vehicles, unmanned ships, intelligent robots with mobile functions, etc. Specifically, the movable platform includes: a body 21, a power system 22, a radar 23, and a control device 24.

The power system 22 is arranged on the body of the machine 21 and used to provide power for the movable platform.

The radar 23 is arranged on the body 21 for collecting point cloud data.

The control device 24 includes a memory 241 and a processor 242.

The memory 241 is used to store computer programs;

The processor 242 is configured to run a computer program stored in the memory to implement:

Further, the processor 242 is further configured to: control the rotation of the radar to sequentially obtain point cloud data corresponding to at least one orientation of the movable platform, and the at least one orientation includes any of the following: front, rear, left ,right.

Further, the processor 242 is further configured to: filter the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data;

Performing clustering processing on the target point cloud data to obtain at least one cluster;

If the number of point cloud data included in the target cluster is greater than or equal to the preset number, it is determined that there is an obstacle in the operating environment, and the target cluster is any one of the at least one cluster.

Further, the processor 242 is further configured to: determine the size information of the obstacle according to the point cloud data in the target cluster;

The movable platform is controlled to avoid obstacles according to the size information of the obstacle.

Further, before determining whether there is an obstacle in the operating environment according to the point cloud data, the processor 242 is further configured to: convert the three-dimensional coordinates of the point cloud data from a spherical coordinate system to the movable platform In the body coordinate system;

The three-dimensional coordinates are converted from the body coordinate system to the inertial coordinate system to determine whether there is an obstacle in the operating environment according to the converted point cloud data.

Further, before controlling the movable platform to avoid obstacles, the processor 242 is further configured to: check whether there are obstacles in the operating environment according to the point cloud data and the movement speed of the movable platform. Test.

Further, the processor 242 is further configured to: according to the point cloud data collected by the radar in the first number of collection periods, determine that the obstacle is different from the movable platform in the first number of collection periods. The first distance value between;

According to the moving speed of the movable platform and the point cloud data collected by the radar in the second number of acquisition periods, it is predicted that the obstacle will be different from the movable platform in the first number of acquisition periods. A second distance value between the two, the first number is greater than the second number;

Check whether there is an obstacle in the operating environment according to the first distance value and the second distance value.

Further, the processor 242 is further configured to: generate a first track according to the first distance value;

Generating a second track according to the second distance value;

If the similarity between the first track and the second track meets a preset range, it is determined that there is an obstacle in the operating environment.

Further, the processor 242 is further configured to: if the similarity between the first trajectory and the second trajectory does not meet a preset range, determine that there is no obstacle in the operating environment;

Delete the first track and the second track;

Control the movable platform to move along the current direction of movement.

Further, the radar 23 is a millimeter wave radar.

Further, the rotation axis of the radar 23 is parallel to the yaw axis of the movable platform.

Further, the radar 23 can continuously rotate 360 degrees around the rotation axis, and collect point cloud data during the continuous rotation.

Further, the radar 23 can intermittently rotate around the rotation axis, the radar stays for a preset time when rotating to a preset angle, and collects point cloud data within the preset time.

The movable platform shown in FIG. 8 can execute the method of the embodiment shown in FIG. 1 to FIG. 6c. For parts that are not described in detail in this embodiment, please refer to the related description of the embodiment shown in FIG. 1 to FIG. 6c. For the implementation process and technical effects of this technical solution, please refer to the description in the embodiment shown in FIG. 1 to FIG. 6c, which will not be repeated here.

In a possible design, the structure of the control device shown in FIG. 9 can be implemented as an electronic device, which can be a controller or a remote server configured in a movable platform. As shown in FIG. 9, the electronic device may include: one or more processors 31 and one or more memories 32. Wherein, the memory 32 is used to store a program that supports the electronic device to execute the obstacle avoidance method provided in the embodiments shown in FIGS. 1 to 6c. The processor 31 is configured to execute a program stored in the memory 32.

Specifically, the program includes one or more computer instructions, and the following steps can be implemented when one or more computer instructions are executed by the processor 31:

Wherein, the structure of the control device may also include a communication interface 33 for the electronic device to communicate with other devices or a communication network.

Further, the processor 31 may be configured to execute: control the rotation of the radar to sequentially obtain point cloud data corresponding to at least one orientation of the movable platform, and the at least one orientation includes any of the following: front, rear, about.

Wherein, the movable platform includes any one of unmanned aerial vehicles, autonomous vehicles, and intelligent robots with mobile functions. The radar is a millimeter wave radar, and the rotation axis of the radar is parallel to the heading axis of the movable platform. The radar can continuously rotate 360 degrees around the rotation axis and collect point cloud data during the continuous rotation, or the radar can intermittently rotate around the rotation axis, and the radar stays for a preset time when rotated to a preset angle, And collect point cloud data within the preset time.

Further, the processor 31 may be configured to perform: screening the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data;

Further, the processor 31 may be configured to execute: determine the size information of the obstacle according to the point cloud data in the target cluster;

Further, the processor 31 may be configured to execute: transform the three-dimensional coordinates of the point cloud data from the spherical coordinate system to the body coordinate system of the movable platform;

Further, the processor 31 may be configured to execute: check whether there is an obstacle in the operating environment according to the point cloud data and the movement speed of the movable platform.

Further, the processor 31 may be configured to execute: according to the point cloud data collected by the radar in a first number of collection periods, determine that the obstacle is different from the movable object in the first number of collection periods. The first distance value between the platforms;

Further, the processor 31 may be configured to execute: generate a first track according to the first distance value;

Generating a second track according to the second distance value;

Further, the processor 31 may be configured to execute: if the similarity between the first trajectory and the second trajectory does not meet a preset range, determine that there is no obstacle in the operating environment;

Delete the first track and the second track;

Control the movable platform to move along the current direction of movement.

The device shown in FIG. 9 can execute the method of the embodiment shown in FIG. 1 to FIG. 6c. For parts that are not described in detail in this embodiment, please refer to the related description of the embodiment shown in FIG. 1 to FIG. 6c. For the implementation process and technical effects of this technical solution, please refer to the description in the embodiment shown in FIG. 1 to FIG. 6c, which will not be repeated here.

In addition, an embodiment of the present invention provides a computer-readable storage medium. The storage medium is a computer-readable storage medium. The computer-readable storage medium stores program instructions. Barrier method.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions recorded in the foregoing embodiments can still be modified, or some or all of the technical features can be equivalently replaced; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention. Scope.

Claims

An obstacle avoidance method, characterized in that it is applied to a movable platform, and the method includes:

Controlling the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment where the movable platform is located;

Determine whether there are obstacles in the operating environment according to the point cloud data;

If there are obstacles in the operating environment, the movable platform is controlled to avoid obstacles.
The method according to claim 1, wherein the controlling the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment of the movable platform comprises:

The rotation of the radar is controlled to sequentially obtain point cloud data corresponding to at least one orientation of the movable platform, and the at least one orientation includes any one of the following: front, back, left, and right.
The method according to claim 1, wherein the determining whether there is an obstacle in the operating environment according to the point cloud data comprises:

Screening the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data;

Performing clustering processing on the target point cloud data to obtain at least one cluster;

If the number of point cloud data included in the target cluster is greater than or equal to the preset number, it is determined that there is an obstacle in the operating environment, and the target cluster is any one of the at least one cluster.
The method according to claim 3, wherein the controlling the movable platform to avoid obstacles comprises:

Determining the size information of the obstacle according to the point cloud data in the target cluster;

The movable platform is controlled to avoid obstacles according to the size information of the obstacle.
The method according to claim 1, wherein before determining whether there is an obstacle in the operating environment according to the point cloud data, the method further comprises:

Transforming the three-dimensional coordinates of the point cloud data from the spherical coordinate system to the body coordinate system of the movable platform;

The three-dimensional coordinates are converted from the body coordinate system to the inertial coordinate system to determine whether there is an obstacle in the operating environment according to the converted point cloud data.
The method according to claim 1, wherein before said controlling said movable platform to avoid obstacles, said method further comprises:

According to the point cloud data and the moving speed of the movable platform, it is checked whether there are obstacles in the operating environment.
The method according to claim 6, wherein the checking whether there is an obstacle in the operating environment according to the point cloud data and the movement speed of the movable platform comprises:

Determine, according to the point cloud data collected by the radar in the first number of collection periods, a first distance value between the obstacles and the movable platform in the first number of collection periods;

According to the moving speed of the movable platform and the point cloud data collected by the radar in the second number of acquisition periods, it is predicted that the obstacle will be different from the movable platform in the first number of acquisition periods. A second distance value between the two, the first number is greater than the second number;

Check whether there is an obstacle in the operating environment according to the first distance value and the second distance value.
The method according to claim 7, wherein the checking whether there is an obstacle in the operating environment according to the first distance value and the second distance value comprises:

Generating a first track according to the first distance value;

Generating a second track according to the second distance value;

If the similarity between the first track and the second track meets a preset range, it is determined that there is an obstacle in the operating environment.
The method according to claim 8, wherein the method further comprises:

If the similarity between the first trajectory and the second trajectory does not meet the preset range, it is determined that there is no obstacle in the operating environment;

Delete the first track and the second track;

Control the movable platform to move along the current direction of movement.
The method according to any one of claims 1 to 9, wherein the movable platform includes any one of an unmanned aerial vehicle, an autonomous driving vehicle, and an intelligent robot with a mobile function.
The method according to any one of claims 1 to 9, wherein the radar is a millimeter wave radar.
The method according to any one of claims 1 to 9, wherein the rotation axis of the radar is parallel to the yaw axis of the movable platform.
The method according to any one of claims 1 to 9, wherein the radar can continuously rotate 360 degrees around a rotation axis, and collect point cloud data during the continuous rotation.
The method according to any one of claims 1 to 9, wherein the radar is capable of intermittently rotating around a rotation axis, and the radar stays for a preset time when rotating to a preset angle, and stops at the preset angle. Set point cloud data to be collected within time.
A movable platform, characterized in that the platform includes: a body, a power system, a radar, and a control device;

The power system is arranged on the body and used to provide power for the movable platform;

The radar is arranged on the body and is used to collect point cloud data;

The control device includes a memory and a processor;

The memory is used to store a computer program;

The processor is configured to run a computer program stored in the memory to realize:

Controlling the rotation of the radar configured on the movable platform to obtain the omnidirectional point cloud data in the operating environment where the movable platform is located;

Determine whether there are obstacles in the operating environment according to the point cloud data;

If there are obstacles in the operating environment, the movable platform is controlled to avoid obstacles.
The movable platform according to claim 15, wherein the processor is further configured to: control the rotation of the radar to sequentially obtain point cloud data corresponding to at least one orientation of the movable platform, and the at least An orientation includes any of the following: front, back, left, and right.
The movable platform according to claim 15, wherein the processor is further configured to: filter the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data; Clustering the target point cloud data to obtain at least one cluster;

If the number of point cloud data included in the target cluster is greater than or equal to the preset number, it is determined that there is an obstacle in the operating environment, and the target cluster is any one of the at least one cluster.
The mobile platform according to claim 17, wherein the processor is further configured to: determine the size information of the obstacle according to the point cloud data in the target cluster;

The movable platform is controlled to avoid obstacles according to the size information of the obstacle.
The movable platform according to claim 15, wherein the processor is further configured to: transform the three-dimensional coordinates of the point cloud data from a spherical coordinate system to the body coordinate system of the movable platform;

The three-dimensional coordinates are converted from the body coordinate system to the inertial coordinate system to determine whether there is an obstacle in the operating environment according to the converted point cloud data.
The movable platform according to claim 15, wherein the processor is further configured to calibrate whether there are obstacles in the operating environment according to the point cloud data and the movement speed of the movable platform. Test.
The movable platform according to claim 21, wherein the processor is further configured to: determine the point cloud data collected by the radar in the first number of collection periods A first distance value between the obstacles and the movable platform in a period;

According to the moving speed of the movable platform and the point cloud data collected by the radar in the second number of acquisition periods, it is predicted that the obstacle will be different from the movable platform in the first number of acquisition periods. A second distance value between the two, the first number is greater than the second number;

Check whether there is an obstacle in the operating environment according to the first distance value and the second distance value.
The mobile platform according to claim 21, wherein the processor is further configured to: generate a first track according to the first distance value;

Generating a second track according to the second distance value;

If the similarity between the first track and the second track meets a preset range, it is determined that there is an obstacle in the operating environment.
The mobile platform according to claim 22, wherein the processor is further configured to: determine if the similarity between the first track and the second track does not meet a preset range There are no obstacles in the operating environment;

Delete the first track and the second track;

Control the movable platform to move along the current direction of movement.
The movable platform according to any one of claims 15 to 23, wherein the movable platform includes any one of an unmanned aerial vehicle, an autonomous driving vehicle, and an intelligent robot with a mobile function.
The movable platform according to any one of claims 15 to 23, wherein the radar is a millimeter wave radar.
The movable platform according to any one of claims 15 to 23, wherein the rotation axis of the radar is parallel to the yaw axis of the movable platform.
The movable platform according to any one of claims 15 to 23, wherein the radar can continuously rotate 360 degrees around a rotation axis, and collect point cloud data during the continuous rotation.
The movable platform according to any one of claims 15 to 23, wherein the radar is capable of intermittently rotating around a rotation axis, and the radar stays for a preset time when rotating to a preset angle, and stays there The point cloud data is collected within the preset time.
A control device, characterized in that the obstacle avoidance device includes:

Memory, used to store computer programs;

The processor is configured to run a computer program stored in the memory to realize:

Controlling the rotation of the radar configured on the movable platform to obtain omnidirectional point cloud data in the operating environment where the movable platform is located;

Determine whether there are obstacles in the operating environment according to the point cloud data;

If there are obstacles in the operating environment, the movable platform is controlled to avoid obstacles.
The device according to claim 29, wherein the processor is further configured to: control the rotation of the radar to sequentially obtain point cloud data corresponding to at least one orientation of the movable platform, and the at least one orientation Including any of the following: front, back, left, right;

Wherein, the movable platform includes any one of unmanned aerial vehicles, autonomous vehicles, and intelligent robots with mobile functions; the radar is a millimeter wave radar; the rotation axis of the radar is parallel to the movable The heading axis of the platform, the radar can continuously rotate 360 degrees around the rotation axis and collect point cloud data during the continuous rotation; or the radar can intermittently rotate around the rotation axis, and the radar can rotate to a preset angle Stay for a preset time, and collect point cloud data within the preset time.
The device according to claim 29, wherein the processor is further configured to: filter the target point cloud data corresponding to the safe movement range of the movable platform from the point cloud data;

Performing clustering processing on the target point cloud data to obtain at least one cluster;

If the number of point cloud data included in the target cluster is greater than or equal to the preset number, it is determined that there is an obstacle in the operating environment, and the target cluster is any one of the at least one cluster.
The device according to claim 31, wherein the processor is further configured to: determine the size information of the obstacle according to the point cloud data in the target cluster;

The movable platform is controlled to avoid obstacles according to the size information of the obstacle.
The device according to claim 29, wherein the processor is further configured to: transform the three-dimensional coordinates of the point cloud data from a spherical coordinate system to the body coordinate system of the movable platform;

The three-dimensional coordinates are converted from the body coordinate system to the inertial coordinate system to determine whether there is an obstacle in the operating environment according to the converted point cloud data.
The device according to claim 29, wherein the processor is further configured to check whether there is an obstacle in the operating environment according to the point cloud data and the movement speed of the movable platform.
The device according to claim 34, wherein the processor is further configured to: determine that within the first number of collection periods according to the point cloud data collected by the radar in the first number of collection periods A first distance value between the obstacle and the movable platform;

According to the moving speed of the movable platform and the point cloud data collected by the radar in the second number of acquisition periods, it is predicted that the obstacle will be different from the movable platform in the first number of acquisition periods. A second distance value between the two, the first number is greater than the second number;

Check whether there is an obstacle in the operating environment according to the first distance value and the second distance value.
The device according to claim 35, wherein the processor is further configured to: generate a first track according to the first distance value;

Generating a second track according to the second distance value;

If the similarity between the first track and the second track meets a preset range, it is determined that there is an obstacle in the operating environment.
The device according to claim 36, wherein the processor is further configured to: if the similarity between the first trajectory and the second trajectory does not meet a preset range, determine the There are no obstacles in the operating environment;

Delete the first track and the second track;

Control the movable platform to move along the current direction of movement.
A computer-readable storage medium, wherein the storage medium is a computer-readable storage medium in which program instructions are stored, and the program instructions are used to implement any one of claims 1 to 14 The obstacle avoidance method described in item.