Anti-collision control method, device and system for remote control car
Through image sensors and target detection models, remote control vehicles can identify obstacles and independently formulate anti-collision strategies, solving the problems of increasing weight of traditional solutions and inability to avoid obstacles, achieving more efficient anti-collision control, and improving the safety of remote control vehicles and user experience.
Patent Information
- Application Number
- CN202411086817.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The anti-collision scheme of traditional remote control vehicles mainly relies on structural improvements, resulting in increased weight and bloated size, and the inability to effectively avoid collisions, especially in complex environments, it is difficult to achieve intelligent and active obstacle avoidance.
Using image sensors and target detection models, we can capture environmental images in real time, identify obstacles and formulate anti-collision control strategies, including parking, detours and overturning, and calculate deceleration, turning and acceleration parameters through distance measuring sensors to realize autonomous anti-collision of remote-controlled vehicles. control.
It improves the safety and operation convenience of remote control vehicles, reduces user manual intervention, provides a more reliable and efficient anti-collision solution, and enhances driving safety and adaptability in complex environments.
Smart Images

Figure CN118915771A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of remote control vehicles, and in particular to an anti-collision control method, device and system for a remote control vehicle. Background Art
[0002] In the development of remote control car technology, in order to improve its driving safety and obstacle avoidance capabilities in complex environments, early engineers mainly focused on improving the structure of remote control cars. This traditional anti-collision solution focuses on resisting potential collision impacts by strengthening the physical structure of the remote control car and reducing damage caused by collisions.
[0003] Specifically, traditional structural improvement methods include but are not limited to strengthening the body frame, adding anti-collision buffer materials, designing retractable or deformable parts, etc. These measures have improved the crash resistance and survivability of the remote control car to a certain extent, allowing the remote control car to better protect its internal mechanical structure and electronic components from damage when it encounters an unexpected collision.
[0004] However, although the structural improvement scheme has improved the anti-collision performance of the remote control car to a certain extent, it still faces many challenges and limitations. Structural improvements often come at the expense of increasing the weight and volume of the remote control car, which may affect the flexibility and maneuverability of the remote control car and reduce its driving ability. Moreover, structural improvements cannot fundamentally solve the problem of how the remote control car can actively avoid collisions during driving. In a rapidly changing or dynamically complex environment, it is far from enough to rely solely on enhanced physical structures to resist collisions. The remote control car needs to have a more intelligent and active obstacle avoidance capability to perceive and avoid potential obstacles in advance.
[0005] Therefore, how to more effectively perform collision avoidance control for remote-controlled vehicles to overcome the limitations of traditional solutions has become a problem that the industry is constantly exploring. Summary of the invention
[0006] In view of this, the embodiments of the present application provide a collision avoidance control method, device and system for a remote-controlled vehicle. The image sensor can capture a wider range of environmental images, and combined with the target detection model, it can achieve comprehensive perception of the surrounding environment, thereby autonomously performing collision avoidance control, reducing the user's manual intervention and attention distraction, improving the remote-controlled vehicle's operational convenience and user experience, and providing a more reliable and efficient solution for the safe driving of the remote-controlled vehicle.
[0007] According to one aspect of the present application, a collision avoidance control method for a remote control vehicle is provided, the method comprising:
[0008] Acquire an environment image captured by an image sensor disposed on the remote control vehicle, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located, wherein the target information includes target speed information, target position information, and target shape information corresponding to at least one target;
[0009] According to the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, determining whether there is an obstacle in each of the targets for the remote control vehicle to travel based on the current driving information, and if there is an obstacle, determining, according to the target to which the obstacle belongs, the target form information corresponding to the target as the obstacle form information of the obstacle;
[0010] An anti-collision control strategy of the remote-controlled vehicle is determined based on the obstacle morphology information, and anti-collision control of the remote-controlled vehicle is performed based on the anti-collision control strategy.
[0011] In an embodiment of the present application, optionally, determining the anti-collision control strategy of the remote control vehicle based on the obstacle morphology information includes:
[0012] Determining the obstacle type to which the obstacle belongs according to the obstacle morphology information;
[0013] If the obstacle type is an obstacle that cannot be circumvented, determining the anti-collision control strategy of the remote control vehicle to be a parking anti-collision strategy;
[0014] If the obstacle type is a circumventable obstacle, determining the anti-collision control strategy of the remote control vehicle to be a circumvention anti-collision strategy;
[0015] If the obstacle type is a surmountable obstacle, the anti-collision control strategy of the remote control vehicle is determined to be a surmounting anti-collision strategy.
[0016] In the embodiment of the present application, optionally, the current driving information includes current speed, current direction and current position;
[0017] The anti-collision control of the remote control vehicle based on the anti-collision control strategy includes:
[0018] When the anti-collision control strategy is the parking anti-collision strategy, the distance between the remote control vehicle and the obstacle is obtained based on a distance measuring sensor, and a target deceleration position and a target deceleration parameter are determined based on the distance and the current speed, and the remote control vehicle is controlled to decelerate at the target deceleration parameter when it reaches the target deceleration position, so that the remote control vehicle stops before reaching the obstacle;
[0019] In the case where the anti-collision control strategy is the detour anti-collision strategy, a target travel route of the remote-controlled vehicle is determined according to the current direction and the current position, a distance between the remote-controlled vehicle and the obstacle determined based on a range-finding sensor is obtained, and a target turning position, a target turning angle, and a target turning speed are determined based on the distance and the current speed, the remote-controlled vehicle is controlled to travel to the target turning position to bypass the obstacle at the target turning angle and the target turning speed, and after bypassing the obstacle, the remote-controlled vehicle is controlled to continue traveling at the current direction and the current speed, so that the remote-controlled vehicle resumes traveling on the target travel route after bypassing the obstacle;
[0020] When the anti-collision control strategy is the rollover anti-collision strategy, the distance between the remote control vehicle and the obstacle determined based on the ranging sensor is obtained, and the target acceleration position and target acceleration parameters of the remote control vehicle are determined according to the distance and the current speed, the remote control vehicle is controlled to accelerate at the target acceleration parameter when it reaches the target acceleration position to roll over the obstacle, and the remote control vehicle is controlled to return to the current speed and continue to move after rolling over the obstacle.
[0021] In the embodiment of the present application, optionally, when the anti-collision control strategy is the over-collision anti-collision strategy, before acquiring the distance between the remote control vehicle and the obstacle determined based on the ranging sensor, the method further includes:
[0022] Based on the current speed and the obstacle shape information, determining whether the remote control vehicle traveling at the current speed satisfies the condition of climbing over the obstacle;
[0023] If the condition for climbing over the obstacle is not met, executing the step of obtaining the distance between the remote control vehicle and the obstacle determined based on the distance measuring sensor;
[0024] If the condition for climbing over the obstacle is met, the step of performing anti-collision control on the remote control vehicle based on the anti-collision control strategy is ignored.
[0025] In the embodiment of the present application, optionally, after performing anti-collision control on the remote control vehicle based on the anti-collision control strategy, the method further includes:
[0026] In the case where the anti-collision control strategy is the parking anti-collision strategy, first guidance information is generated according to the current driving information, and parking anti-collision teaching information is generated according to the first guidance information, the target deceleration position and the target deceleration parameter, wherein the parking anti-collision teaching information is used to guide the user to control the remote control vehicle according to the first guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target deceleration position and the target deceleration parameter, so as to realize parking anti-collision operation learning;
[0027] In the case where the anti-collision control strategy is the detour anti-collision strategy, second guidance information is generated according to the current driving information, and detour anti-collision teaching information is generated according to the second guidance information, the target turning position, the target turning angle, and the target turning speed, wherein the detour anti-collision teaching information is used to guide the user to control the remote control vehicle according to the second guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target turning position, the target turning angle, and the target turning speed, so as to achieve detour anti-collision operation learning;
[0028] In the case where the anti-collision control strategy is the rollover anti-collision strategy, third guidance information is generated according to the current driving information, and rollover anti-collision teaching information is generated according to the third guidance information, the target acceleration position and the target acceleration parameter, wherein the rollover anti-collision teaching information is used to guide the user to control the remote control vehicle according to the third guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target acceleration position and the target acceleration parameter to realize rollover anti-collision operation learning.
[0029] In the embodiment of the present application, optionally, the first guidance information includes a first guidance position and a first remote control vehicle control parameter, the first guidance position being a starting position of the remote control vehicle for learning a parking collision avoidance operation, and the first remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the first guidance position to meet the driving condition corresponding to the current driving information;
[0030] The second guidance information includes a second guidance position and a second remote control vehicle control parameter, the second guidance position being a starting position of the remote control vehicle for learning the detour collision avoidance operation, and the second remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the second guidance position to meet the driving condition corresponding to the current driving information;
[0031] The third guidance information includes a third guidance position and a third remote control vehicle control parameter, wherein the third guidance position is the starting position of the remote control vehicle for learning the rollover collision avoidance operation, and the third remote control vehicle control parameter is a control parameter that enables the remote control vehicle stationary at the third guidance position to meet the driving conditions corresponding to the current driving information.
[0032] In the embodiment of the present application, optionally, before acquiring the environment image captured by the image sensor disposed on the remote control vehicle, the method further includes:
[0033] Turning on the automatic collision avoidance function of the remote control car;
[0034] After performing anti-collision control on the remote control vehicle based on the anti-collision control strategy, the method further includes:
[0035] Outputting teaching inquiry information according to the type of the anti-collision strategy to inquire the user whether to perform corresponding type of anti-collision operation learning;
[0036] In the case where the anti-collision strategy is the parking anti-collision strategy, if a feedback signal confirming the parking anti-collision operation learning is received, the remote control car is controlled to automatically drive to the first guidance position to output the parking anti-collision teaching information, and when the user controls the remote control car to drive to the target deceleration position according to the first remote control car control parameter, a first reminder signal is output;
[0037] In the case where the collision avoidance strategy is the detour collision avoidance strategy, if a feedback signal confirming the detour collision avoidance operation learning is received, the remote control vehicle is controlled to automatically drive to the second guiding position to output the detour collision avoidance teaching information, and when the user controls the remote control vehicle to drive to the target turning position according to the second remote control vehicle control parameter, a second reminder signal is output;
[0038] In the case where the anti-collision strategy is the rollover anti-collision strategy, if a feedback signal confirming the learning of the rollover anti-collision operation is received, the remote control car is controlled to automatically drive to the third guidance position to output the rollover anti-collision teaching information, and when the user controls the remote control car to drive to the target acceleration position according to the third remote control car control parameter, a third reminder signal is output.
[0039] In an embodiment of the present application, optionally, the acquiring of an environment image captured by an image sensor disposed on the remote control vehicle, and performing target detection on the environment image through a target detection model to determine target information of the environment in which the remote control vehicle is located, includes:
[0040] Acquire frame by frame the environment image captured by the image sensor disposed on the remote control vehicle, and perform target detection on each frame of the environment image through the target detection model, determine at least one target prediction frame in each frame of the environment image, and determine target position information and target morphology information corresponding to each target according to each target prediction frame;
[0041] Convert each frame reference image corresponding to the environmental image into the coordinates of the environmental image, determine the target prediction frame of each target in the environmental image in the reference image according to the target prediction frame corresponding to the environmental image and each reference image, and determine the target speed information corresponding to each target in the environmental image according to the target prediction frame of each target in the environmental image and the target prediction frame in the reference image, wherein the reference image is a preset frame environmental image before the environmental image.
[0042] According to another aspect of the present application, there is provided an anti-collision control device for a remote control vehicle, the device comprising:
[0043] An acquisition module is used to acquire an environment image collected by an image sensor installed on the remote control vehicle, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located, wherein the target information includes target speed information, target position information and target shape information corresponding to at least one target;
[0044] an identification module, for determining, according to the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, whether there is an obstacle in each of the targets for the remote control vehicle to travel based on the current driving information, and if there is an obstacle, determining, according to the target to which the obstacle belongs, the target form information corresponding to the target as the obstacle form information of the obstacle;
[0045] A control module is used to determine an anti-collision control strategy of the remote control vehicle based on the obstacle morphology information, and perform anti-collision control on the remote control vehicle based on the anti-collision control strategy.
[0046] In the embodiment of the present application, optionally, the control module is further used to:
[0047] Determining the obstacle type to which the obstacle belongs according to the obstacle morphology information;
[0048] If the obstacle type is an obstacle that cannot be circumvented, determining the anti-collision control strategy of the remote control vehicle to be a parking anti-collision strategy;
[0049] If the obstacle type is a circumventable obstacle, determining the anti-collision control strategy of the remote control vehicle to be a circumvention anti-collision strategy;
[0050] If the obstacle type is a surmountable obstacle, the anti-collision control strategy of the remote control vehicle is determined to be a surmounting anti-collision strategy.
[0051] In the embodiment of the present application, optionally, the current driving information includes current speed, current direction and current position; the control module is further used to:
[0052] When the anti-collision control strategy is the parking anti-collision strategy, the distance between the remote control vehicle and the obstacle is obtained based on a distance measuring sensor, and a target deceleration position and a target deceleration parameter are determined based on the distance and the current speed, and the remote control vehicle is controlled to decelerate at the target deceleration parameter when it reaches the target deceleration position, so that the remote control vehicle stops before reaching the obstacle;
[0053] In the case where the anti-collision control strategy is the detour anti-collision strategy, a target travel route of the remote-controlled vehicle is determined according to the current direction and the current position, a distance between the remote-controlled vehicle and the obstacle determined based on a range-finding sensor is obtained, and a target turning position, a target turning angle, and a target turning speed are determined based on the distance and the current speed, the remote-controlled vehicle is controlled to travel to the target turning position to bypass the obstacle at the target turning angle and the target turning speed, and after bypassing the obstacle, the remote-controlled vehicle is controlled to continue traveling at the current direction and the current speed, so that the remote-controlled vehicle returns to the target travel route after bypassing the obstacle;
[0054] When the anti-collision control strategy is the rollover anti-collision strategy, the distance between the remote control vehicle and the obstacle determined based on the ranging sensor is obtained, and the target acceleration position and target acceleration parameters of the remote control vehicle are determined according to the distance and the current speed, the remote control vehicle is controlled to accelerate at the target acceleration parameter when it reaches the target acceleration position to roll over the obstacle, and the remote control vehicle is controlled to return to the current speed and continue to move after rolling over the obstacle.
[0055] In the embodiment of the present application, optionally, the control module is further used to:
[0056] In the case where the anti-collision control strategy is the over-collision anti-collision strategy, judging whether the remote control vehicle traveling at the current speed satisfies the condition of over-crossing the obstacle based on the current speed and the obstacle shape information;
[0057] If the condition for climbing over the obstacle is not met, executing the step of obtaining the distance between the remote control vehicle and the obstacle determined based on the distance measuring sensor;
[0058] If the condition for climbing over the obstacle is met, the step of performing anti-collision control on the remote control vehicle based on the anti-collision control strategy is ignored.
[0059] In the embodiment of the present application, optionally, the control module is further used to:
[0060] In the case where the anti-collision control strategy is the parking anti-collision strategy, first guidance information is generated according to the current driving information, and parking anti-collision teaching information is generated according to the first guidance information, the target deceleration position and the target deceleration parameter, wherein the parking anti-collision teaching information is used to guide the user to control the remote control vehicle according to the first guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target deceleration position and the target deceleration parameter, so as to realize parking anti-collision operation learning;
[0061] In the case where the anti-collision control strategy is the detour anti-collision strategy, second guidance information is generated according to the current driving information, and detour anti-collision teaching information is generated according to the second guidance information, the target turning position, the target turning angle, and the target turning speed, wherein the detour anti-collision teaching information is used to guide the user to control the remote control vehicle according to the second guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target turning position, the target turning angle, and the target turning speed, so as to achieve detour anti-collision operation learning;
[0062] In the case where the anti-collision control strategy is the rollover anti-collision strategy, third guidance information is generated according to the current driving information, and rollover anti-collision teaching information is generated according to the third guidance information, the target acceleration position and the target acceleration parameter, wherein the rollover anti-collision teaching information is used to guide the user to control the remote control vehicle according to the third guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target acceleration position and the target acceleration parameter to realize rollover anti-collision operation learning.
[0063] In the embodiment of the present application, optionally, the first guidance information includes a first guidance position and a first remote control vehicle control parameter, the first guidance position being a starting position of the remote control vehicle for learning a parking collision avoidance operation, and the first remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the first guidance position to meet the driving condition corresponding to the current driving information;
[0064] The second guidance information includes a second guidance position and a second remote control vehicle control parameter, the second guidance position being a starting position of the remote control vehicle for learning the detour collision avoidance operation, and the second remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the second guidance position to meet the driving condition corresponding to the current driving information;
[0065] The third guidance information includes a third guidance position and a third remote control vehicle control parameter, wherein the third guidance position is the starting position of the remote control vehicle for learning the rollover collision avoidance operation, and the third remote control vehicle control parameter is a control parameter that enables the remote control vehicle stationary at the third guidance position to meet the driving conditions corresponding to the current driving information.
[0066] In the embodiment of the present application, optionally, the acquisition module is further used to: enable the automatic collision avoidance function of the remote control car;
[0067] The control module is also used for:
[0068] Outputting teaching inquiry information according to the type of the anti-collision strategy to inquire the user whether to perform corresponding type of anti-collision operation learning;
[0069] In the case where the anti-collision strategy is the parking anti-collision strategy, if a feedback signal confirming the parking anti-collision operation learning is received, the remote control car is controlled to automatically drive to the first guidance position to output the parking anti-collision teaching information, and when the user controls the remote control car to drive to the target deceleration position according to the first remote control car control parameter, a first reminder signal is output;
[0070] In the case where the collision avoidance strategy is the detour collision avoidance strategy, if a feedback signal confirming the detour collision avoidance operation learning is received, the remote control vehicle is controlled to automatically drive to the second guiding position to output the detour collision avoidance teaching information, and when the user controls the remote control vehicle to drive to the target turning position according to the second remote control vehicle control parameter, a second reminder signal is output;
[0071] In the case where the anti-collision strategy is the rollover anti-collision strategy, if a feedback signal confirming the learning of the rollover anti-collision operation is received, the remote control car is controlled to automatically drive to the third guidance position to output the rollover anti-collision teaching information, and when the user controls the remote control car to drive to the target acceleration position according to the third remote control car control parameter, a third reminder signal is output.
[0072] In the embodiment of the present application, optionally, the acquisition module is further used to:
[0073] Acquire frame by frame the environment image captured by the image sensor disposed on the remote control vehicle, and perform target detection on each frame of the environment image through the target detection model, determine at least one target prediction frame in each frame of the environment image, and determine target position information and target morphology information corresponding to each target according to each target prediction frame;
[0074] Convert each frame reference image corresponding to the environmental image into the coordinates of the environmental image, determine the target prediction frame of each target in the environmental image in the reference image according to the target prediction frame corresponding to the environmental image and each reference image, and determine the target speed information corresponding to each target in the environmental image according to the target prediction frame of each target in the environmental image and the target prediction frame in the reference image, wherein the reference image is a preset frame environmental image before the environmental image.
[0075] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the anti-collision control method of the remote control vehicle is implemented.
[0076] According to another aspect of the present application, a collision avoidance control system for a remote control vehicle is provided, comprising a remote control vehicle body and a controller, wherein the controller comprises a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned collision avoidance control method for the remote control vehicle when executing the computer program.
[0077] By means of the above technical scheme, the embodiment of the present application provides a method, device and system for anti-collision control of a remote control car. First, an image sensor is used to collect environmental images in real time, and a target detection model is used to identify the target in the image, and its speed, position and shape information are extracted. Subsequently, according to this information and the current driving state of the remote control car, it is determined whether there is an obstacle and the specific shape of the obstacle is determined. Finally, a corresponding anti-collision control strategy is formulated and executed based on the shape information of the obstacle to ensure the safe driving of the remote control car. Compared with the prior art, the embodiment of the present application mainly improves the structure of the remote control car to achieve anti-collision. Through the image sensor, a wider range of environmental images can be captured. Combined with the target detection model, it can achieve a comprehensive perception of the surrounding environment and reduce the possibility of missed detection. In addition, the method based on image processing and target detection is not affected by environmental factors such as light changes and obstacle occlusion, and can maintain high detection accuracy and stability in complex environments, thereby reducing the weight increase and bulkiness of the remote control car caused by changes in the structure of the remote control car. In addition, since the system can autonomously perform anti-collision control, it reduces the user's manual intervention and distraction, improves the convenience of operation and user experience of the remote control car, and provides a more reliable and efficient solution for the safe driving of the remote control car.
[0078] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0080] Figure 1 A schematic diagram of a flow chart of a collision avoidance control method for a remote control vehicle provided in an embodiment of the present application is shown;
[0081] Figure 2 A schematic flow chart of another anti-collision control method for a remote control vehicle provided in an embodiment of the present application is shown;
[0082] Figure 3 A schematic structural diagram of an anti-collision control device for a remote control vehicle provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0083] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.
[0084] In this embodiment, a collision avoidance control method for a remote control vehicle is provided. Figure 1 As shown, the method includes:
[0085] Step 101, obtain an environment image captured by an image sensor installed on the remote control car, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control car is located, wherein the target information includes target speed information, target position information and target shape information corresponding to at least one target.
[0086] Step 102, based on the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, determine whether there is an obstacle in each of the targets for the remote control vehicle to move based on the current driving information, and if there is an obstacle, based on the target to which the obstacle belongs, determine that the target shape information corresponding to the target is the obstacle shape information of the obstacle.
[0087] Step 103: determining an anti-collision control strategy for the remote control vehicle based on the obstacle morphology information, and performing anti-collision control on the remote control vehicle based on the anti-collision control strategy.
[0088] The anti-collision control method of the remote control car provided in the embodiment of the present application integrates advanced image processing and target detection technology, aims to improve the safety and autonomy of the remote control car during driving, and helps to avoid the damage of the remote control car due to improper operation when children or other users operate the toy remote control car. Among them, an image sensor (such as a camera) is installed on the remote control car to capture images of the surrounding environment in real time. In the process of operating the remote control car, the image sensor continuously collects environmental images around the remote control car, processes these images through a target detection model (such as a deep learning model, a YOLOv8 model, etc.), identifies the targets in the environment, and extracts the speed information, position information and morphological information of each target. The application of the target detection model improves the accuracy and efficiency of target recognition and ensures the rapid recognition of potential obstacles. Then, according to the target speed information, target position information and current driving information (such as speed and direction) of each target, it is judged which targets may constitute obstacles, that is, targets that may affect the continued safe driving of the remote control car, so as to accurately identify obstacles that may pose a threat to the remote control car, avoid unnecessary anti-collision measures, and improve the efficiency and response speed of the system. When it is determined that there is an obstacle, the specific obstacle form is further determined based on the target form information of the obstacle (such as size, shape, whether it is movable, etc.), which helps to formulate a more accurate and effective anti-collision control strategy. Finally, based on the form information of the obstacle, a specific anti-collision control strategy is formulated, such as stopping, detouring, climbing over, etc. By formulating and executing accurate control strategies, the risk of collision caused by human operation errors is reduced. Then, according to the formulated anti-collision control strategy, the control system of the remote control car (specifically the controller of the remote control car) performs corresponding actions to avoid collision with obstacles, realizing autonomous anti-collision control during the driving of the remote control car, improving the safety and reliability of driving, and improving the adaptability and intelligence level of the remote control car.
[0089] By applying the technical solution of this embodiment, firstly, the environment image is collected in real time by the image sensor, and the target in the image is identified by the target detection model, and its speed, position and shape information are extracted. Subsequently, according to this information and the current driving state of the remote control car, it is judged whether there is an obstacle and the specific shape of the obstacle is determined. Finally, the corresponding anti-collision control strategy is formulated and executed based on the shape information of the obstacle to ensure the safe driving of the remote control car. Compared with the prior art, the embodiment of the present application mainly realizes the anti-collision method by improving the structure of the remote control car. The image sensor can capture a wider range of environmental images, and combined with the target detection model, it can realize the comprehensive perception of the surrounding environment and reduce the possibility of missed detection. The method based on image processing and target detection is not affected by environmental factors such as light changes and obstacle occlusion, and can maintain high detection accuracy and stability in complex environments, thereby reducing the weight increase and bulky volume of the remote control car caused by changes in the structure of the remote control car. In addition, since the system can autonomously perform anti-collision control, it reduces the manual intervention and distraction of the user, improves the convenience of operation and user experience of the remote control car, and provides a more reliable and efficient solution for the safe driving of the remote control car.
[0090] In an embodiment of the present application, optionally, step 101 includes: acquiring the environmental image captured by the image sensor set on the remote control car frame by frame, and performing target detection on each frame of the environmental image through the target detection model, determining at least one target prediction frame in each frame of the environmental image, and determining the target position information and target morphology information corresponding to each target according to each target prediction frame; converting each frame of the reference image corresponding to the environmental image into the coordinates of the environmental image, determining the target prediction frame of each target in the environmental image in the reference image according to the environmental image and the target prediction frame corresponding to each reference image, and determining the target speed information corresponding to each target in the environmental image according to the target prediction frame of each target in the environmental image and the target prediction frame in the reference image.
[0091] In this embodiment, the image sensor continuously captures images of the environment around the remote control car and processes them frame by frame to ensure the real-time and continuity of the environmental information. The target detection model is used to detect the target of each frame of the environmental image, and the target prediction frame containing the target in the image is predicted. These target prediction frames not only identify the position of the target, but also imply the target morphological information of the target. The target prediction frame generated by the target detection model can accurately determine the position of each target in the environmental image. At the same time, the shape and size of the target prediction frame also reflect the morphological information of the target. Further processing of the target prediction frame can extract more specific morphological parameters, such as the size and shape of the target, to provide more detailed information for subsequent anti-collision control. Here, the target morphological information can be simply determined according to the edge of the target prediction frame to improve the efficiency of determining the target morphological information, and the image covered by the target prediction frame can also be edge detected to determine the target morphological information. Afterwards, the coordinate system of each frame reference image (previous preset frame environmental image) corresponding to the environmental image is converted to the coordinate system of the current environmental image to compare and track cross-frame targets. By comparing the target prediction frames of the corresponding targets in the environment image and each reference image, the moving trajectory of the target between different frames can be tracked, and then the speed information of the target can be calculated according to the position change of the target in the continuous frames. This speed estimation method based on image sequence is more accurate and reliable than the speed estimation of a single frame, and the calculation is simple and efficient. Through the calculation scheme of the embodiment of the present application, the environment image is processed frame by frame, and combined with the target detection model, the target in the surrounding environment can be identified in real time and accurately, providing reliable data support for subsequent anti-collision control, not only determining the position information of the target, but also extracting the morphological information of the target through the shape and size of the target prediction frame, providing a basis for formulating a more accurate anti-collision control strategy, in addition, through cross-frame comparison and coordinate conversion, the tracking of the target moving trajectory and the estimation of the speed information are realized, which provides the possibility for predicting the future position of the target and dynamically adjusting the anti-collision control strategy. And the method based on image processing and target detection is not limited by factors such as ambient light and occlusion, and can maintain high detection accuracy and stability in complex and changeable environments, improving the intelligence and adaptability of the system.
[0092] In the embodiment of the present application, optionally, the current driving information includes current speed, current direction and current position; step 103 includes:
[0093] Step 103-1: Determine the obstacle type of the obstacle according to the obstacle shape information.
[0094] In this embodiment, based on the intelligent judgment of the obstacle morphology information and combined with the ranging sensor data, the remote control car adopts different anti-collision control strategies for different types of obstacles, thereby realizing precise anti-collision control for different scenarios. First, based on the obstacle morphology information determined by image processing and target detection technology, the system can intelligently judge the type of obstacle (non-circumventable obstacles, circumventable obstacles, traversable obstacles), providing a basis for the subsequent selection of appropriate anti-collision strategies.
[0095] Step 103-2, if the obstacle type is an obstacle that cannot be circumvented, determine that the collision avoidance control strategy of the remote control vehicle is a parking collision avoidance strategy; obtain the distance between the remote control vehicle and the obstacle determined based on the ranging sensor, and determine the target deceleration position and target deceleration parameters based on the distance and the current speed, and control the remote control vehicle to decelerate at the target deceleration parameter when it reaches the target deceleration position, so that the remote control vehicle stops before reaching the obstacle.
[0096] In this embodiment, when the obstacle is an obstacle that cannot be circumvented, a parking anti-collision strategy is selected. Specifically, if the size of the obstacle is determined to exceed a preset size (including a preset height and a preset length) based on the obstacle morphology information, then the obstacle can be considered to be an obstacle that cannot be circumvented, such as a wall. In the case of determining to select a parking anti-collision strategy, the target deceleration position and target deceleration parameters are calculated in combination with the distance measured by the ranging sensor and the current speed of the remote control car. Subsequently, it is switched to automatic control, and the remote control car continues to be controlled to move at the current speed and current direction. When the remote control car reaches the target deceleration position, it decelerates with the target deceleration parameters until it stops completely, thereby ensuring that the remote control car can park safely when encountering an obstacle that cannot be circumvented, avoiding possible collision accidents, protecting the safety of the remote control car, and avoiding damage after a collision.
[0097] Step 103-3, if the obstacle type is a circumventable obstacle, determine that the collision avoidance control strategy of the remote control vehicle is a circumvention collision avoidance strategy; determine the target travel route of the remote control vehicle according to the current direction and the current position, obtain the distance between the remote control vehicle and the obstacle determined based on the ranging sensor, and determine the target turning position, target turning angle and target turning speed based on the distance and the current speed, control the remote control vehicle to move to the target turning position to bypass the obstacle at the target turning angle and the target turning speed, and control the remote control vehicle to continue moving at the current direction and the current speed after bypassing the obstacle, so that the remote control vehicle returns to the target travel route after bypassing the obstacle.
[0098] In this embodiment, when the obstacle is a circumventable obstacle, a detour anti-collision strategy is selected. Specifically, if the height of the obstacle exceeds the preset height but the length is less than the preset length, then the obstacle can be considered to be a circumventable obstacle, such as a table leg. In the case of determining to select the detour anti-collision strategy, the target route that the user operates the remote control car to expect it to travel is determined according to the current direction and current position, and the target turning position, target turning angle and target turning speed are calculated in combination with the distance detected by the ranging sensor and the current speed. Then, it is switched to automatic control, and the remote control car is continued to be controlled to travel at the current speed and current direction. When the remote control car reaches the target turning position, it turns according to the target turning angle and target turning speed. After bypassing the obstacle, the original direction and speed (i.e., the current direction and current speed in the current driving information) are restored, and the original desired target route is returned, thereby improving the flexibility and adaptability of the remote control car when facing circumventable obstacles, and ensuring the continuity and safety of driving.
[0099] Step 103-4, if the obstacle type is a surmountable obstacle, determine that the collision avoidance control strategy of the remote control vehicle is a surmounting collision avoidance strategy; obtain the distance between the remote control vehicle and the obstacle determined based on the ranging sensor, and determine the target acceleration position and target acceleration parameters of the remote control vehicle according to the distance and the current speed, control the remote control vehicle to accelerate at the target acceleration parameter when it reaches the target acceleration position to surmount the obstacle, and control the remote control vehicle to return to the current speed and continue moving after surmounting the obstacle.
[0100] In this embodiment, when the obstacle is a surmountable obstacle, a surmounting anti-collision strategy is selected. Specifically, if the height of the obstacle does not exceed the preset height, then the obstacle can be considered to be a surmountable obstacle, such as a small slope. In the case of determining to select the surmounting anti-collision strategy, the target acceleration position and target acceleration parameters are calculated based on the distance measured by the ranging sensor and the current speed. Subsequently, the remote control is switched to automatic control to continue to control the remote control car to move at the current speed and current direction. When the remote control car reaches the target acceleration position, it is accelerated according to the target acceleration parameters, and the obstacle is surmounted at a sufficient speed, and the original speed is restored after surmounting to continue moving. For surmountable obstacles, the direct surmounting method not only simplifies the driving path, but also improves driving efficiency. At the same time, the smoothness and safety of the surmounting process are ensured by precise acceleration control. By applying the technical solution of this embodiment, a suitable anti-collision strategy is automatically selected according to the obstacle type, thereby realizing an intelligent decision-making process. Different anti-collision strategies are used to ensure the safe driving of the remote control car in front of different types of obstacles, while improving the driving efficiency. In addition, the distance detected by the ranging sensor and the current driving information are combined to calculate accurate deceleration, turning and acceleration parameters, thereby realizing precise control of the remote control car and improving the safety of the remote control car.
[0101] In an embodiment of the present application, optionally, before obtaining the distance between the remote control car and the obstacle determined based on the ranging sensor in step 103-4, the method further includes: judging whether the remote control car traveling at the current speed satisfies the condition of climbing over the obstacle based on the current speed and the obstacle morphology information; if the condition of climbing over the obstacle is not met, executing the step of obtaining the distance between the remote control car and the obstacle determined based on the ranging sensor; if the condition of climbing over the obstacle is met, ignoring the step of performing anti-collision control on the remote control car based on the anti-collision control strategy.
[0102] In this embodiment, for the anti-collision strategy of step 103-4, a judgment is added on whether the current speed of the remote control car meets the conditions for climbing over the obstacle. This improvement ensures that the remote control car has sufficient speed and power conditions before trying to climb over the obstacle, thereby improving the success rate and safety of the climbing process. Among them, before obtaining the distance between the remote control car and the obstacle, firstly judge based on the current speed and obstacle morphological information (such as height). If the current speed is not enough to support the remote control car to climb over the obstacle (for example, the speed is too slow, which may lead to insufficient power and failure to successfully climb over), then continue to execute the subsequent steps, that is, obtain the distance and prepare to implement the anti-collision strategy for climbing over. If the current speed has met or exceeded the conditions for climbing over the obstacle (that is, the speed is fast enough and the power is sufficient), the system determines that the remote control car can directly climb over the obstacle without further anti-collision control. That is, if the climbing condition is met, the system will ignore the step of anti-collision control of the remote control car based on the anti-collision control strategy, and directly allow the remote control car to continue to move at the current speed or a slightly adjusted speed to climb over the obstacle, reducing unnecessary control intervention and improving driving efficiency. In addition, since the speed condition has been judged, the safety of the climbing process is also guaranteed. If the climbing condition is not met, the system will execute according to the original anti-collision strategy for climbing, that is, obtain the distance between the remote control car and the obstacle, and calculate the target acceleration position and target acceleration parameters based on this information, and control the remote control car to accelerate to climb over the obstacle. The embodiment of the present application can avoid climbing failures caused by insufficient speed by pre-judging whether the speed of the remote control car meets the climbing condition, thereby improving the success rate of the climbing process. For situations where the climbing condition has been met, the system allows the remote control car to climb over directly, reducing unnecessary control steps and improving driving efficiency.
[0103] In the embodiment of the present application, optionally, the above-mentioned anti-collision control method of the remote control car can be executed only under the premise that the automatic anti-collision function of the remote control car is turned on. Further, the anti-collision control method of the remote control car can be used to teach the user about anti-collision operation, such as Figure 2 As shown, after step 103, the following steps are also included:
[0104] Step 201, when the anti-collision control strategy is the parking anti-collision strategy, first guidance information is generated according to the current driving information, and parking anti-collision teaching information is generated according to the first guidance information, the target deceleration position and the target deceleration parameter, wherein the parking anti-collision teaching information is used to guide the user to control the remote control vehicle according to the first guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target deceleration position and the target deceleration parameter to realize parking anti-collision operation learning.
[0105] In the embodiment of the present application, not only the automatic anti-collision control of the remote control car is realized, but also this control method is further used to teach the user anti-collision operation. This teaching function not only enhances the intelligence and practicality of the remote control car, but also helps to improve the user's experience and skill level. In the case of executing the parking anti-collision strategy, instructions or suggestions for guiding the user to control the remote control car are generated according to the current driving information (such as current speed, current direction, current position), that is, the first guidance information, which are intended to help the user adjust the remote control car to a state suitable for executing the parking anti-collision strategy. Combined with the first guidance information, the target deceleration position and the target deceleration parameter, the parking anti-collision teaching information is generated to guide the user how to control the remote control car to decelerate and stop in front of the obstacle to avoid collision, so that through simulation and guidance, the user can master how to correctly control the remote control car to perform parking anti-collision operations when encountering obstacles that cannot be bypassed.
[0106] Step 202, when the anti-collision control strategy is the detour anti-collision strategy, second guidance information is generated according to the current driving information, and detour anti-collision teaching information is generated according to the second guidance information, the target turning position, the target turning angle and the target turning speed, wherein the detour anti-collision teaching information is used to guide the user to control the remote control vehicle according to the second guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target turning position, the target turning angle and the target turning speed to realize detour anti-collision operation learning.
[0107] In the embodiment of the present application, when executing the detour collision avoidance strategy, similarly, based on the current driving information, an instruction or suggestion to guide the user to detour the obstacle, i.e., the second guidance information, is generated to help the user adjust the remote control car to a state suitable for executing the detour collision avoidance strategy. In combination with the second guidance information, the target turning position, the target turning angle, and the target turning speed, detour collision avoidance teaching information is generated to guide the user on how to control the remote control car to detour the obstacle according to the predetermined route and parameters, so that through simulation and guidance, the user can master how to flexibly control the remote control car to perform detour collision avoidance operations when encountering detourable obstacles.
[0108] Step 203, when the anti-collision control strategy is the rollover anti-collision strategy, third guidance information is generated according to the current driving information, and rollover anti-collision teaching information is generated according to the third guidance information, the target acceleration position and the target acceleration parameter, wherein the rollover anti-collision teaching information is used to guide the user to control the remote control vehicle according to the third guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target acceleration position and the target acceleration parameter to realize rollover anti-collision operation learning.
[0109] In the embodiment of the present application, when the overtaking collision avoidance strategy is executed, similarly, based on the current driving information, instructions or suggestions are generated to guide the user to prepare to overtake the obstacle, i.e., the third guidance information, to help the user adjust the remote control car to a state suitable for executing the overtaking collision avoidance strategy. In combination with the third guidance information, the target acceleration position and the target acceleration parameter, overtaking collision avoidance teaching information is generated to guide the user on how to control the remote control car to accelerate and successfully overtake the obstacle, so that through simulation and guidance, the user can master how to control the remote control car to accelerate and overtake when encountering an obstacle that can be overtaken.
[0110] Step 204: output teaching inquiry information according to the type of the anti-collision strategy to inquire the user whether to perform corresponding type of anti-collision operation learning.
[0111] In the embodiment of the present application, after executing the anti-collision strategy to realize the anti-collision control of the remote control car, an inquiry is added as to whether the user is willing to learn the anti-collision operation, so that the corresponding type of teaching process can be automatically executed according to the user's feedback signal, and a real-time reminder signal is provided to help the user better master the operation skills. By outputting the teaching inquiry information to ask the user whether he wants to learn the corresponding type of anti-collision operation, this step increases the user's participation, allowing the user to choose whether to learn according to his own needs, and improves the pertinence and effectiveness of the teaching.
[0112] Step 205: When the anti-collision strategy is the parking anti-collision strategy, if a feedback signal confirming the parking anti-collision operation learning is received, the remote control car is controlled to automatically drive to the first guide position to output the parking anti-collision teaching information, and when the user controls the remote control car to drive to the target deceleration position according to the first remote control car control parameter, a first reminder signal is output. The first guide information includes a first guide position and a first remote control car control parameter, the first guide position is the starting position of the remote control car for the parking anti-collision operation learning, and the first remote control car control parameter is a control parameter that enables the remote control car stationary at the first guide position to meet the driving conditions corresponding to the current driving information.
[0113] In the embodiment of the present application, after the anti-collision strategy is determined to be a parking anti-collision strategy and the user confirms to conduct this type of learning, the remote control car automatically drives to the first guide position, which is the starting point for the user to learn the parking anti-collision operation. After reaching the first guide position, the system outputs parking anti-collision teaching information to guide the user on how to operate the remote control car to perform parking anti-collision. Specifically, the user should be taught to first control the remote control car according to the first remote control car control parameter (such as speed, direction, etc.), and then when the remote control car reaches the target deceleration position, the remote control car is controlled to decelerate based on the target deceleration parameter so as to stop before reaching the obstacle. In order to further allow the user to better grasp the timing of controlling the remote control car to decelerate, when the user controls the remote control car to drive to the target deceleration position (or close to the position, such as the first distance from the position), the system outputs a first reminder signal to prompt the user to start the deceleration operation at this time.
[0114] Step 206, when the anti-collision strategy is the detour anti-collision strategy, if a feedback signal confirming the detour anti-collision operation learning is received, the remote control car is controlled to automatically drive to the second guide position to output the detour anti-collision teaching information, and when the user controls the remote control car to drive to the target turning position according to the second remote control car control parameter, a second reminder signal is output. The second guide information includes the second guide position and the second remote control car control parameter, the second guide position is the starting position of the remote control car for detour anti-collision operation learning, and the second remote control car control parameter is the control parameter that enables the remote control car stationary at the second guide position to meet the driving conditions corresponding to the current driving information.
[0115] In an embodiment of the present application, when the anti-collision strategy is to bypass anti-collision and the user confirms the learning, the learning process starts. The remote control car automatically drives to the second guide position as the starting point for learning the bypass anti-collision operation. At this position, the system outputs the bypass anti-collision teaching information, and instructs the user in detail how to bypass the obstacle. Specifically, the user should be taught to first control the remote control car according to the second remote control car control parameter, and then when the remote control car reaches the target turning position, the remote control car is controlled to turn based on the target turning angle and target turning speed to bypass the obstacle. In order to further allow users to better control the turning timing of the remote control car, when the remote control car reaches the target turning position (or approaches the position, such as the second distance from the position), the system outputs a second reminder signal to prompt the user to prepare to turn.
[0116] Step 207, when the anti-collision strategy is the over-the-top anti-collision strategy, if a feedback signal confirming the over-the-top anti-collision operation learning is received, the remote control car is controlled to automatically drive to the third guide position to output the over-the-top anti-collision teaching information, and when the user controls the remote control car to drive to the target acceleration position according to the third remote control car control parameter, a third reminder signal is output. The third guide information includes a third guide position and a third remote control car control parameter, the third guide position is the starting position of the remote control car for the over-the-top anti-collision operation learning, and the third remote control car control parameter is a control parameter that enables the remote control car stationary at the third guide position to meet the driving conditions corresponding to the current driving information.
[0117] In an embodiment of the present application, after the anti-collision strategy is determined to be over-collision prevention and the user agrees to learn, the learning process starts. The remote control car automatically drives to the third guide position, which is the starting position for learning the over-collision prevention operation. At this position, over-collision prevention teaching information is output to explain how to control the remote control car to overturn obstacles. Specifically, the user should be taught to first control the remote control car according to the third remote control car control parameter, and then when the remote control car drives to the target acceleration position, the remote control car is accelerated based on the target acceleration parameter so that it can overturn the obstacle. In order to further allow the user to better grasp the timing of controlling the acceleration of the remote control car, when the user controls the remote control car to drive to the target acceleration position (or when it is close to the position, for example, the third distance from the position), the system outputs a third reminder signal to prompt the user to accelerate to overturn the obstacle.
[0118] By applying the technical solution of this embodiment, through anti-collision operation teaching, users can systematically learn and master the anti-collision control skills of remote control cars and improve their driving level, so that after users master the correct anti-collision operation method, they can control the remote control car more safely during actual use and reduce the occurrence of collision accidents. Combining the automatic anti-collision control function of the remote control car with the user teaching function not only realizes intelligent control, but also improves the practicality and educational value of the product, making the remote control car not only a simple toy or tool, but also a partner who can learn and grow, enhancing the user's sense of participation and satisfaction. In addition, the anti-collision teaching information can be directly generated using the anti-collision control parameters of the remote control car generated by the user in the process of operating the remote control car, so that the anti-collision teaching is not only more suitable for the scene where the user uses the remote control car, so that the user can learn at any time according to local conditions, and the user can directly learn in the scene where he is, master the operation skills under different anti-collision strategies, thereby improving the safety and coping ability in actual driving.
[0119] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides an anti-collision control device for a remote control car, such as Figure 3 As shown, the device comprises:
[0120] The acquisition module 301 is used to acquire an environment image collected by an image sensor installed on the remote control vehicle, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located, wherein the target information includes target speed information, target position information and target shape information corresponding to at least one target;
[0121] The identification module 302 is used to determine whether there is an obstacle for the remote control vehicle to travel based on the current driving information in each of the targets according to the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, and if there is an obstacle, determine the target shape information corresponding to the target as the obstacle shape information of the obstacle according to the target to which the obstacle belongs;
[0122] The control module 303 is used to determine the anti-collision control strategy of the remote control vehicle based on the obstacle morphology information, and perform anti-collision control on the remote control vehicle based on the anti-collision control strategy.
[0123] In the embodiment of the present application, optionally, the control module 303 is further configured to:
[0124] Determining the obstacle type to which the obstacle belongs according to the obstacle morphology information;
[0125] If the obstacle type is an obstacle that cannot be circumvented, determining the anti-collision control strategy of the remote control vehicle to be a parking anti-collision strategy;
[0126] If the obstacle type is a circumventable obstacle, determining the anti-collision control strategy of the remote control vehicle to be a circumvention anti-collision strategy;
[0127] If the obstacle type is a surmountable obstacle, the anti-collision control strategy of the remote control vehicle is determined to be a surmounting anti-collision strategy.
[0128] In the embodiment of the present application, optionally, the current driving information includes current speed, current direction and current position; the control module 303 is further used to:
[0129] When the anti-collision control strategy is the parking anti-collision strategy, the distance between the remote control vehicle and the obstacle is obtained based on a distance measuring sensor, and a target deceleration position and a target deceleration parameter are determined based on the distance and the current speed, and the remote control vehicle is controlled to decelerate at the target deceleration parameter when it reaches the target deceleration position, so that the remote control vehicle stops before reaching the obstacle;
[0130] In the case where the anti-collision control strategy is the detour anti-collision strategy, a target travel route of the remote-controlled vehicle is determined according to the current direction and the current position, a distance between the remote-controlled vehicle and the obstacle determined based on a range-finding sensor is obtained, and a target turning position, a target turning angle, and a target turning speed are determined based on the distance and the current speed, the remote-controlled vehicle is controlled to travel to the target turning position to bypass the obstacle at the target turning angle and the target turning speed, and after bypassing the obstacle, the remote-controlled vehicle is controlled to continue traveling at the current direction and the current speed, so that the remote-controlled vehicle returns to the target travel route after bypassing the obstacle;
[0131] When the anti-collision control strategy is the rollover anti-collision strategy, the distance between the remote control vehicle and the obstacle determined based on the ranging sensor is obtained, and the target acceleration position and target acceleration parameters of the remote control vehicle are determined according to the distance and the current speed, the remote control vehicle is controlled to accelerate at the target acceleration parameter when it reaches the target acceleration position to roll over the obstacle, and the remote control vehicle is controlled to return to the current speed and continue to move after rolling over the obstacle.
[0132] In the embodiment of the present application, optionally, the control module 303 is further configured to:
[0133] In the case where the anti-collision control strategy is the over-collision anti-collision strategy, judging whether the remote control vehicle traveling at the current speed satisfies the condition of over-crossing the obstacle based on the current speed and the obstacle shape information;
[0134] If the condition for climbing over the obstacle is not met, executing the step of obtaining the distance between the remote control vehicle and the obstacle determined based on the distance measuring sensor;
[0135] If the condition for climbing over the obstacle is met, the step of performing anti-collision control on the remote control vehicle based on the anti-collision control strategy is ignored.
[0136] In the embodiment of the present application, optionally, the control module 303 is further configured to:
[0137] In the case where the anti-collision control strategy is the parking anti-collision strategy, first guidance information is generated according to the current driving information, and parking anti-collision teaching information is generated according to the first guidance information, the target deceleration position and the target deceleration parameter, wherein the parking anti-collision teaching information is used to guide the user to control the remote control vehicle according to the first guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target deceleration position and the target deceleration parameter, so as to realize parking anti-collision operation learning;
[0138] In the case where the anti-collision control strategy is the detour anti-collision strategy, second guidance information is generated according to the current driving information, and detour anti-collision teaching information is generated according to the second guidance information, the target turning position, the target turning angle, and the target turning speed, wherein the detour anti-collision teaching information is used to guide the user to control the remote control vehicle according to the second guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target turning position, the target turning angle, and the target turning speed, so as to achieve detour anti-collision operation learning;
[0139] In the case where the anti-collision control strategy is the rollover anti-collision strategy, third guidance information is generated according to the current driving information, and rollover anti-collision teaching information is generated according to the third guidance information, the target acceleration position and the target acceleration parameter, wherein the rollover anti-collision teaching information is used to guide the user to control the remote control vehicle according to the third guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target acceleration position and the target acceleration parameter to realize rollover anti-collision operation learning.
[0140] In the embodiment of the present application, optionally, the first guidance information includes a first guidance position and a first remote control vehicle control parameter, the first guidance position being a starting position of the remote control vehicle for learning a parking collision avoidance operation, and the first remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the first guidance position to meet the driving condition corresponding to the current driving information;
[0141] The second guidance information includes a second guidance position and a second remote control vehicle control parameter, the second guidance position being a starting position of the remote control vehicle for learning the detour collision avoidance operation, and the second remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the second guidance position to meet the driving condition corresponding to the current driving information;
[0142] The third guidance information includes a third guidance position and a third remote control vehicle control parameter, wherein the third guidance position is the starting position of the remote control vehicle for learning the rollover collision avoidance operation, and the third remote control vehicle control parameter is a control parameter that enables the remote control vehicle stationary at the third guidance position to meet the driving conditions corresponding to the current driving information.
[0143] In the embodiment of the present application, optionally, the acquisition module 301 is further used to: enable the automatic collision avoidance function of the remote control car;
[0144] The control module 303 is further used for:
[0145] Outputting teaching inquiry information according to the type of the anti-collision strategy to inquire the user whether to perform corresponding type of anti-collision operation learning;
[0146] In the case where the anti-collision strategy is the parking anti-collision strategy, if a feedback signal confirming the parking anti-collision operation learning is received, the remote control car is controlled to automatically drive to the first guidance position to output the parking anti-collision teaching information, and when the user controls the remote control car to drive to the target deceleration position according to the first remote control car control parameter, a first reminder signal is output;
[0147] In the case where the collision avoidance strategy is the detour collision avoidance strategy, if a feedback signal confirming the detour collision avoidance operation learning is received, the remote control vehicle is controlled to automatically drive to the second guiding position to output the detour collision avoidance teaching information, and when the user controls the remote control vehicle to drive to the target turning position according to the second remote control vehicle control parameter, a second reminder signal is output;
[0148] In the case where the anti-collision strategy is the rollover anti-collision strategy, if a feedback signal confirming the learning of the rollover anti-collision operation is received, the remote control car is controlled to automatically drive to the third guidance position to output the rollover anti-collision teaching information, and when the user controls the remote control car to drive to the target acceleration position according to the third remote control car control parameter, a third reminder signal is output.
[0149] In the embodiment of the present application, optionally, the acquisition module 301 is further used to:
[0150] Acquire frame by frame the environment image captured by the image sensor disposed on the remote control vehicle, and perform target detection on each frame of the environment image through the target detection model, determine at least one target prediction frame in each frame of the environment image, and determine target position information and target morphology information corresponding to each target according to each target prediction frame;
[0151] Convert each frame reference image corresponding to the environmental image into the coordinates of the environmental image, determine the target prediction frame of each target in the environmental image in the reference image according to the target prediction frame corresponding to the environmental image and each reference image, and determine the target speed information corresponding to each target in the environmental image according to the target prediction frame of each target in the environmental image and the target prediction frame in the reference image, wherein the reference image is a preset frame environmental image before the environmental image.
[0152] It should be noted that for other corresponding descriptions of the functional units involved in the anti-collision control device for a remote control vehicle provided in the embodiment of the present application, reference can be made to Figure 1 to Figure 2 The corresponding description in the method will not be repeated here.
[0153] The present application also provides a collision avoidance control system for a remote control vehicle, including a remote control vehicle body and a controller, wherein the controller includes a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the computer program is executed by the processor, the steps in each method embodiment are implemented.
[0154] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.
[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.
[0156] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.
[0157] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0158] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A collision avoidance control method for a remote control vehicle, characterized in that: The method comprises: Acquire an environment image captured by an image sensor disposed on the remote control vehicle, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located, wherein the target information includes target speed information, target position information, and target shape information corresponding to at least one target; According to the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, determining whether there is an obstacle in each of the targets for the remote control vehicle to travel based on the current driving information, and if there is an obstacle, determining, according to the target to which the obstacle belongs, the target form information corresponding to the target as the obstacle form information of the obstacle; An anti-collision control strategy of the remote-controlled vehicle is determined based on the obstacle morphology information, and anti-collision control of the remote-controlled vehicle is performed based on the anti-collision control strategy.
2. The method according to claim 1, characterized in that The step of determining the anti-collision control strategy of the remote control vehicle based on the obstacle morphology information includes: Determining the obstacle type to which the obstacle belongs according to the obstacle morphology information; If the obstacle type is an obstacle that cannot be circumvented, determining the anti-collision control strategy of the remote control vehicle to be a parking anti-collision strategy; If the obstacle type is a circumventable obstacle, determining the anti-collision control strategy of the remote control vehicle to be a circumvention anti-collision strategy; If the obstacle type is a surmountable obstacle, the anti-collision control strategy of the remote control vehicle is determined to be a surmounting anti-collision strategy.
3. The method according to claim 2, characterized in that The current driving information includes current speed, current direction and current position; The anti-collision control of the remote control vehicle based on the anti-collision control strategy includes: When the anti-collision control strategy is the parking anti-collision strategy, the distance between the remote control vehicle and the obstacle is obtained based on a distance measuring sensor, and a target deceleration position and a target deceleration parameter are determined based on the distance and the current speed, and the remote control vehicle is controlled to decelerate at the target deceleration parameter when it reaches the target deceleration position, so that the remote control vehicle stops before reaching the obstacle; In the case where the anti-collision control strategy is the detour anti-collision strategy, a target travel route of the remote-controlled vehicle is determined according to the current direction and the current position, a distance between the remote-controlled vehicle and the obstacle determined based on a range-finding sensor is obtained, and a target turning position, a target turning angle, and a target turning speed are determined based on the distance and the current speed, the remote-controlled vehicle is controlled to travel to the target turning position to bypass the obstacle at the target turning angle and the target turning speed, and after bypassing the obstacle, the remote-controlled vehicle is controlled to continue traveling at the current direction and the current speed, so that the remote-controlled vehicle resumes traveling on the target travel route after bypassing the obstacle; When the anti-collision control strategy is the rollover anti-collision strategy, the distance between the remote control vehicle and the obstacle determined based on the ranging sensor is obtained, and the target acceleration position and target acceleration parameters of the remote control vehicle are determined according to the distance and the current speed, the remote control vehicle is controlled to accelerate at the target acceleration parameter when it reaches the target acceleration position to roll over the obstacle, and the remote control vehicle is controlled to return to the current speed and continue to move after rolling over the obstacle.
4. The method according to claim 3, characterized in that In the case where the anti-collision control strategy is the over-collision anti-collision strategy, before acquiring the distance between the remote control vehicle and the obstacle determined based on the ranging sensor, the method further includes: Based on the current speed and the obstacle shape information, determining whether the remote control vehicle traveling at the current speed satisfies the condition of climbing over the obstacle; If the condition for climbing over the obstacle is not met, executing the step of obtaining the distance between the remote control vehicle and the obstacle determined based on the distance measuring sensor; If the condition for climbing over the obstacle is met, the step of performing anti-collision control on the remote control vehicle based on the anti-collision control strategy is ignored.
5. The method according to claim 3, characterized in that: After performing anti-collision control on the remote control vehicle based on the anti-collision control strategy, the method further includes: In the case where the anti-collision control strategy is the parking anti-collision strategy, first guidance information is generated according to the current driving information, and parking anti-collision teaching information is generated according to the first guidance information, the target deceleration position and the target deceleration parameter, wherein the parking anti-collision teaching information is used to guide the user to control the remote control vehicle according to the first guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target deceleration position and the target deceleration parameter, so as to realize parking anti-collision operation learning; In the case where the anti-collision control strategy is the detour anti-collision strategy, second guidance information is generated according to the current driving information, and detour anti-collision teaching information is generated according to the second guidance information, the target turning position, the target turning angle, and the target turning speed, wherein the detour anti-collision teaching information is used to guide the user to control the remote control vehicle according to the second guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target turning position, the target turning angle, and the target turning speed, so as to achieve detour anti-collision operation learning; In the case where the anti-collision control strategy is the rollover anti-collision strategy, third guidance information is generated according to the current driving information, and rollover anti-collision teaching information is generated according to the third guidance information, the target acceleration position and the target acceleration parameter, wherein the rollover anti-collision teaching information is used to guide the user to control the remote control vehicle according to the third guidance information so that the remote control vehicle meets the driving conditions corresponding to the current driving information, and to guide the user to control the remote control vehicle based on the target acceleration position and the target acceleration parameter to realize rollover anti-collision operation learning.
6. The method according to claim 5, characterized in that The first guidance information includes a first guidance position and a first remote control vehicle control parameter, wherein the first guidance position is a starting position of the remote control vehicle for learning parking collision avoidance operation, and the first remote control vehicle control parameter is a control parameter that enables the remote control vehicle stationary at the first guidance position to meet the driving condition corresponding to the current driving information; The second guidance information includes a second guidance position and a second remote control vehicle control parameter, the second guidance position being a starting position of the remote control vehicle for learning the detour collision avoidance operation, and the second remote control vehicle control parameter being a control parameter that enables the remote control vehicle stationary at the second guidance position to meet the driving condition corresponding to the current driving information; The third guidance information includes a third guidance position and a third remote control vehicle control parameter, wherein the third guidance position is the starting position of the remote control vehicle for learning the rollover collision avoidance operation, and the third remote control vehicle control parameter is a control parameter that enables the remote control vehicle stationary at the third guidance position to meet the driving conditions corresponding to the current driving information.
7. The method according to claim 6, characterized in that Before acquiring the environment image captured by the image sensor disposed on the remote control vehicle, the method further includes: Turning on the automatic collision avoidance function of the remote control car; After performing anti-collision control on the remote control vehicle based on the anti-collision control strategy, the method further includes: Outputting teaching inquiry information according to the type of the anti-collision strategy to inquire the user whether to perform corresponding type of anti-collision operation learning; In the case where the anti-collision strategy is the parking anti-collision strategy, if a feedback signal confirming the parking anti-collision operation learning is received, the remote control car is controlled to automatically drive to the first guidance position to output the parking anti-collision teaching information, and when the user controls the remote control car to drive to the target deceleration position according to the first remote control car control parameter, a first reminder signal is output; In the case where the collision avoidance strategy is the detour collision avoidance strategy, if a feedback signal confirming the detour collision avoidance operation learning is received, the remote control vehicle is controlled to automatically drive to the second guiding position to output the detour collision avoidance teaching information, and when the user controls the remote control vehicle to drive to the target turning position according to the second remote control vehicle control parameter, a second reminder signal is output; In the case where the anti-collision strategy is the rollover anti-collision strategy, if a feedback signal confirming the learning of the rollover anti-collision operation is received, the remote control car is controlled to automatically drive to the third guidance position to output the rollover anti-collision teaching information, and when the user controls the remote control car to drive to the target acceleration position according to the third remote control car control parameter, a third reminder signal is output.
8. The method according to any one of claims 1 to 7, characterized in that The step of acquiring an environment image captured by an image sensor disposed on the remote control vehicle and performing target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located includes: Acquire frame by frame the environment image captured by the image sensor disposed on the remote control vehicle, and perform target detection on each frame of the environment image through the target detection model, determine at least one target prediction frame in each frame of the environment image, and determine target position information and target morphology information corresponding to each target according to each target prediction frame; Convert each frame reference image corresponding to the environmental image into the coordinates of the environmental image, determine the target prediction frame of each target in the environmental image in the reference image according to the target prediction frame corresponding to the environmental image and each reference image, and determine the target speed information corresponding to each target in the environmental image according to the target prediction frame of each target in the environmental image and the target prediction frame in the reference image, wherein the reference image is a preset frame environmental image before the environmental image.
9. A collision prevention control device for a remote control vehicle, characterized in that: The device comprises: An acquisition module is used to acquire an environment image collected by an image sensor installed on the remote control vehicle, and perform target detection on the environment image through a target detection model to determine target information of the environment where the remote control vehicle is located, wherein the target information includes target speed information, target position information and target shape information corresponding to at least one target; an identification module, for determining, according to the target speed information, the target position information and the current driving information corresponding to the remote control vehicle, whether there is an obstacle in each of the targets for the remote control vehicle to travel based on the current driving information, and if there is an obstacle, determining, according to the target to which the obstacle belongs, the target form information corresponding to the target as the obstacle form information of the obstacle; A control module is used to determine an anti-collision control strategy of the remote control vehicle based on the obstacle morphology information, and perform anti-collision control on the remote control vehicle based on the anti-collision control strategy.
10. A collision avoidance control system for a remote control vehicle, comprising a remote control vehicle body and a controller, wherein the controller comprises a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.
Citation Information
Patent Citations
Obstacle prompt method and device based on user motion state and terminal
CN107133572A
Robot obstacle avoidance control method and system, and robot
CN112650235A
Method and device for preventing collision of automatic driving vehicle and electronic equipment
CN114475660A
Obstacle crossing method and device of unmanned vehicle, unmanned vehicle and storage medium
CN115214635A
Vehicle guiding method and device, vehicle and storage medium
CN116166011A
