Intelligent driving obstacle avoidance processing method, electronic equipment, storage medium and computer program product

Abstract

The invention discloses an intelligent driving obstacle avoidance processing method, which comprises the following steps of: 1, judging whether an obstacle in front of the driving direction of an intelligent driving vehicle is in a planned driving path of the intelligent driving vehicle or not, if not, continuing to drive according to the planned driving path, and if so, entering a step 2; and 2, detecting the relative distance between the intelligent driving vehicle and an obstacle in real time, detecting a lane line to judge whether a lane is changed or not once the relative distance is smaller than a set braking safety distance, executing a lane changing obstacle avoidance decision if so, further calculating a lane changing vehicle distance, executing a lane changing action if the relative distance is larger than the set lane changing safety distance, and executing a deceleration obstacle avoidance decision if not. The invention also discloses electronic equipment, a storage medium and a computer program product. According to the method, the traditional main consideration of the speed factor is converted into the main consideration of the distance factor, the adaptability to dynamic obstacles and static obstacles is improved, the collision risk is dynamically evaluated, the accuracy of a decision result is improved, and the purposes of improving the driving efficiency and safety are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology. More specifically, the present invention relates to an intelligent driving obstacle avoidance processing method, an electronic device, a storage medium and a computer program product. Background Art

[0002] Intelligent driving technology is a technology that uses machines to assist or replace human driving, which can greatly improve the efficiency of urban traffic operations. Ensuring driving safety is a basic requirement of intelligent driving technology, and safe obstacle avoidance through environmental perception is an important part of intelligent driving decision-making and control.

[0003] Traditional intelligent driving obstacle avoidance uses sensors to collect environmental information within the external detectable range, and determines whether to actively avoid obstacles based on rules. It is necessary to preset rules in advance based on manual experience and calibration work. There may be a lot of information that needs to be processed at the same time and lack of effectiveness, making it difficult for intelligent driving cars to make decisions in time, affecting driving efficiency and safety. Summary of the invention

[0004] The present invention provides an intelligent driving obstacle avoidance processing method and electronic equipment, storage medium and computer program product, which change the traditional main consideration of speed factor to the main consideration of distance factor, improve the adaptability to dynamic obstacles and static obstacles, dynamically evaluate the collision risk, improve the accuracy of decision results, and thus improve the purpose of driving efficiency and safety.

[0005] The present invention provides an intelligent driving obstacle avoidance processing method and electronic equipment, storage medium and computer program product, which change the traditional main consideration of speed factor to the main consideration of distance factor, improve the adaptability to dynamic obstacles and static obstacles, dynamically evaluate the collision risk, improve the accuracy of decision results, and thus improve the purpose of driving efficiency and safety.

[0006] In order to achieve these purposes and other advantages according to the present invention, an intelligent driving obstacle avoidance processing method is provided, comprising: Step 1: determine whether the obstacle in front of the intelligent driving vehicle is within the planned driving path of the intelligent driving vehicle. If not, continue to drive according to the planned driving path. If yes, proceed to step 2. Step 2: Detect the relative distance between the intelligent driving vehicle and the obstacle in real time. Once it is less than the set braking safety distance, detect the lane line to determine whether the lane can be changed. If it is, execute the lane change obstacle avoidance decision, and further calculate the variable lane vehicle distance. If it is greater than the set lane change safety distance, execute the lane change action; otherwise, execute the deceleration obstacle avoidance decision.

[0007] Preferably, the set braking safety distance is twice the minimum braking distance, and the minimum braking distance is calculated as follows: Where S represents the minimum braking distance, v 0 Represents the speed of the intelligent driving vehicle, m Represents the friction coefficient between the intelligent driving vehicle and the road surface, g represents the acceleration due to gravity, α represents the correction factor, obtained through empirical data, w 实载 Represents the actual load weight of the smart driving vehicle, w 空载 Represents the unladen weight of the smart driving vehicle, β Represents the response time of the intelligent driving vehicle braking command system, c Represents the error factor, obtained through empirical data.

[0008] Preferably, the set lane change safety distance is twice the minimum lane change distance, and the calculation of the minimum lane change distance is achieved by constructing a BP neural network. Specifically, a data set is constructed, including the longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the target lane risk vehicle, the lane change time, the actual load weight of the intelligent driving vehicle, the empty weight of the intelligent driving vehicle, and the lane change distance of the intelligent driving vehicle. The data set is divided into a training set, a validation set, and a test set, and a BP neural network is constructed, wherein the longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the target lane risk vehicle, the lane change time, the actual load weight of the intelligent driving vehicle, and the empty weight of the intelligent driving vehicle are used as the input layer of the BP neural network, and the lane change distance of the intelligent driving vehicle is used as the output layer of the BP neural network. The number of hidden layers is set to 10, and the advantages and disadvantages of the BP neural network are judged according to the correlation and error between the predicted value and the measured value, and the minimum lane change distance is predicted based on the BP neural network.

[0009] Preferably, the obstacle directly in front of the smart driving vehicle is determined by radar and camera. The radar is used to detect the relative distance between the smart driving vehicle and the obstacle directly in front of the smart driving vehicle and the relative distance between the smart driving vehicle and the target lane risk vehicles in the lanes to the left and right of the smart driving vehicle and the direction of its travel. The camera is used to detect lane lines to determine whether lanes can be changed and to photograph the driving environment of the smart driving vehicle.

[0010] Preferably, the frequency of the radar acquiring the distance information is 10 Hz. Once the distance is less than a set braking safety distance, the frequency of the radar acquiring the distance information is 100 Hz.

[0011] Preferably, the obstacle is judged to be a static obstacle or a dynamic obstacle. If it is a static obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time. If it is a dynamic obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time, and the speed of the intelligent driving vehicle is detected in real time. According to the speed of the obstacle, the relative distance between the intelligent driving vehicle and the obstacle is calculated in real time, and the smaller value of the detected relative distance and the calculated relative distance is taken as the relative distance between the intelligent driving vehicle and the obstacle for subsequent steps.

[0012] An electronic device comprises: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the described method.

[0013] A storage medium stores a computer program, which, when executed by a processor, implements the method described.

[0014] A computer program product comprises a computer program, which, when executed by a processor, implements the method described.

[0015] The present invention has at least the following beneficial effects: First, the present invention detects the relative distance between the intelligent driving vehicle and the obstacle at different times, changes the traditional main consideration of speed factor to the main consideration of distance factor, improves the adaptability to dynamic obstacles and static obstacles, thereby ensuring the appropriateness and effectiveness of the response, dynamically evaluates the collision risk by comparing the relative distance and the braking safety distance, and automatically takes measures. By sequentially detecting whether the intelligent driving vehicle has the physical conditions for changing lanes and whether the intelligent driving vehicle has the threat factors for changing lanes under the condition of changing lanes, a solution is provided for the safe obstacle avoidance of the intelligent driving vehicle in a complex dynamic environment, so as to achieve the purpose of improving the accuracy of the decision-making results, and then improving the driving efficiency and safety.

[0016] Second, in order to implement safeguard measures, the present invention sets the braking safety distance to twice the minimum braking distance, and the lane change safety distance to twice the minimum lane change distance, so as to facilitate early warning and decision-making of intelligent driving vehicles. The calculation of the minimum braking distance takes into account road conditions, vehicle quality, performance of the vehicle's braking system, etc. The minimum lane change distance constructs a BP neural network to facilitate early warning and decision-making of intelligent driving vehicles, and better predict and classify the minimum lane change distance of intelligent driving vehicles during actual driving.

[0017] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. DETAILED DESCRIPTION

[0018] The present invention is further described in detail below with reference to the description so that those skilled in the art can implement it accordingly.

[0019] It should be understood that the terms such as “having”, “including” and “comprising” used herein do not exclude the existence or addition of one or more other elements or combinations thereof.

[0020] It should be noted that the experimental methods described in the following embodiments are conventional methods unless otherwise specified, and the reagents and materials can be obtained from commercial channels unless otherwise specified.

[0021] The present invention provides an intelligent driving obstacle avoidance processing method, comprising: Step 1: The intelligent driving vehicle plans a complete driving path according to the destination, and detects obstacles on the current path through components such as radar and camera. For example, the detection range is 50m ahead, and it is determined whether the obstacle in front of the intelligent driving vehicle is within the planned driving path of the intelligent driving vehicle. Otherwise, it continues to drive according to the planned driving path. If yes, it proceeds to step 2. Step 2: Real-time detection of the relative distance between the intelligent driving vehicle and the obstacle as the only parameter requirement for executing different response measures. Once it is less than the set braking safety distance, which can be set to 20 m, the surrounding environment information of the intelligent driving vehicle is collected, and the lane line is detected to determine whether the lane can be changed. If so, a lane change and obstacle avoidance decision is executed, and the variable lane distance is further calculated. If it is greater than the set lane change safety distance, which can be set to 15 m, the lane change action is executed, otherwise the deceleration and obstacle avoidance decision is executed.

[0022] In the above technical scheme, by detecting the relative distance between the intelligent driving vehicle and the obstacle at different times, the main consideration factor is changed from the traditional speed factor to the distance factor, so as to improve the adaptability to dynamic obstacles and static obstacles, thereby ensuring the appropriateness and effectiveness of the response. By comparing the relative distance and the braking safety distance, the collision risk is dynamically evaluated and measures are taken automatically. By successively detecting whether the intelligent driving vehicle has the physical conditions for changing lanes and whether the intelligent driving vehicle has the threat factors for changing lanes under the conditions of changing lanes, a solution is provided for the safe obstacle avoidance of the intelligent driving car in a complex dynamic environment, so as to improve the accuracy of the decision-making results, and then improve the driving efficiency and safety.

[0023] In another technical solution, the set braking safety distance is twice the minimum braking distance, and the minimum braking distance is calculated as follows: Where S represents the minimum braking distance, v 0Represents the speed of the intelligent driving vehicle, m Represents the friction coefficient between the intelligent driving vehicle and the road surface, g represents the acceleration due to gravity, α represents the correction factor, obtained through empirical data, w 实载 Represents the actual load weight of the smart driving vehicle, w 空载 Represents the unladen weight of the smart driving vehicle, β Represents the response time of the intelligent driving vehicle braking command system, c Represents the error factor, obtained through empirical data.

[0024] In the above technical solution, once the relative distance between the intelligent driving vehicle and the obstacle in front reaches the minimum braking distance, there is a great possibility of collision. In order to implement the safeguard measure, the braking safety distance is set to twice the minimum braking distance, so as to facilitate the early warning and decision-making of the intelligent driving vehicle. In actual situations, many factors should be considered, such as road conditions, vehicle quality, performance of the vehicle braking system, etc., taking into account the different friction coefficients of intelligent driving vehicles on different road surfaces, and also taking into account the impact of the load of intelligent driving vehicles on the braking of intelligent driving vehicles, the above minimum braking distance calculation method is formed, and the correction factor α The response time of the intelligent driving vehicle braking command system can be calculated based on the historical data of the intelligent driving vehicle. β It can be calculated based on the historical data of intelligent driving vehicles, and the error factor c Can be obtained through experience.

[0025] In another technical solution, the set lane change safety distance is twice the minimum lane change distance. The calculation of the minimum lane change distance is realized by constructing a BP neural network, so that the signal propagates forward from the input layer to the hidden layer to the output layer, and the error propagates backward from the output layer to the hidden layer to the input layer, and the weights and biases from the hidden layer to the output layer are adjusted in turn. Specifically, a data set is constructed, including the longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the risk vehicle in the target lane, the lane change time, the actual load weight of the intelligent driving vehicle, and the intelligent driving The empty weight of the moving vehicle, the lane changing distance of the intelligent driving vehicle, and the selection of input layer parameters depend on the establishment of the virtual scene and the establishment of the simulation working conditions. Comprehensive consideration is given to the intelligent driving vehicles with different loads at different speeds (acceleration, deceleration, and stationary) when preparing to change lanes. The possibility of collision of the target lane risk vehicle at different speeds (acceleration, deceleration, and stationary) is obtained. Specifically, the collision and non-collision (lane changing time) are obtained, and the factors related to the lane changing distance are determined according to the simulation results. Only in this way can the model effectively reflect the relationship between the relevant factors of the intelligent driving vehicle and the target lane risk vehicle and the lane changing distance. The data set is divided into a training set, a validation set, and a test set, for example, 70%, 15%, and 15%, and a BP neural network is constructed, wherein the longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the target lane risk vehicle, the lane change time, the actual load weight of the intelligent driving vehicle, and the unloaded weight of the intelligent driving vehicle are used as the input layer of the BP neural network, and the lane change distance of the intelligent driving vehicle is used as the output layer of the BP neural network. The number of hidden layers is trained from 1 to 20, and the number of hidden layers is set to 10 according to the training of different numbers of hidden layers. It continues to converge during 15 iterations, and the performance loss function of the fitting error test of the training set and the validation set continues to decrease. R The value is 0.9934, and the total R value is 0.9921. The relationship between the predicted value and the measured value is close, which meets the test accuracy requirements. That is, the constructed BP neural network can judge the quality of the BP neural network according to the correlation and error between the predicted value and the measured value, and the minimum lane change distance can be predicted based on the BP neural network.

[0026] In the above technical solution, in order to implement the safeguard measures, that is, the lane change safety distance is set to twice the minimum lane change distance, so as to facilitate the early warning and decision-making of the intelligent driving vehicle, by constructing a BP neural network and utilizing its powerful model fitting ability and good generalization ability, the minimum lane change distance of the intelligent driving vehicle in the actual driving process can be better predicted and classified.

[0027] In another technical solution, the obstacle directly in front of the intelligent driving vehicle is determined by radar and camera. The radar is used to detect the relative distance between the intelligent driving vehicle and the obstacle directly in front of it in its driving direction, and is also used to detect the relative distance between the intelligent driving vehicle and the target lane risk vehicles in the lanes on the left and right sides of it in its driving direction. The radar can be preferably at the millimeter level and can be set around the intelligent driving vehicle. It can detect distance data, direction data, speed data, etc. The camera is used to detect lane lines to determine whether lanes can be changed, and is also used to photograph the driving environment of the intelligent driving vehicle. It can be set around the intelligent driving vehicle. The radar and camera form the perception layer of the intelligent driving vehicle, which is used to detect the obstacle recognition and photographing the driving environment of the intelligent driving vehicle, and cooperate with each other to improve the recognition accuracy to complete high-precision road condition monitoring and recognition. The data transmitted by the radar and camera is the data basis for forming the logical analysis calculation and decision-making of the intelligent driving vehicle.

[0028] In another technical solution, the radar obtains distance information at a frequency of 10Hz. Once it is less than the set braking safety distance, the radar obtains distance information at a frequency of 100Hz. During safe driving, intelligent driving vehicles can extract features at a lower frequency. The longer the response cycle, the higher the frequency of feature extraction. Once in a dangerous driving process, the shorter the response cycle, the priority collection and response of environmental information for emergency vehicle conditions can be achieved, further improving the effectiveness of environmental perception and the safety of intelligent driving.

[0029] In another technical solution, it is determined whether the obstacle is a static obstacle or a dynamic obstacle. If it is a static obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time. If it is a dynamic obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time, and the speed of the intelligent driving vehicle is detected in real time. According to the speed of the obstacle, the relative distance between the intelligent driving vehicle and the obstacle is calculated in real time. The smaller value of the detected relative distance and the calculated relative distance is taken as the relative distance between the intelligent driving vehicle and the obstacle for subsequent steps.

[0030] In the above technical solution, the obstacle is judged as a static obstacle or a dynamic obstacle by speed detection. The relative distance of the static obstacle depends on the intelligent driving vehicle, and the relative distance of the dynamic obstacle depends on the dynamic obstacle and the intelligent driving vehicle. By comparing the detected relative distance with the calculated relative distance, a small distance is crucial for the intelligent driving vehicle to make a decision under emergency driving conditions. By taking a small value, the relative distance is further limited, the collision risk is reduced, and the accuracy of the decision is improved.

[0031] The present invention also provides an electronic device, comprising: at least one processor, and a memory connected to the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the above method. The electronic device can be any terminal device including a mobile phone, a laptop computer, a desktop computer, a tablet computer, a PDA (Personal Digital Assistant), a POS (Point of Sales), a vehicle-mounted computer, etc.

[0032] The present invention also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above method is implemented.

[0033] Through the description of the above implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present invention, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, RandomAccess Memory), a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0034] The present invention also provides a computer program product, which includes a computer program. When the program is executed by a processor, the above method is implemented.

[0035] A computer program (which may also be referred to or described as a program, software, software application, module, software module, script or code) may be written in any form of programming language, including compiled or interpreted languages ​​or declarative or procedural languages, and the computer program may be deployed in any form, including as a standalone program or as a module, component, subroutine or other unit suitable for use in a computing environment. A computer program may, but need not, correspond to a file in a file system. A program may be stored in a portion of a file that stores other programs or data, for example, one or more scripts stored in the following: in a markup language document; in a single file dedicated to a related program; or in multiple collaborative files, for example, files storing one or more modules, subroutines or code portions. A computer program may be deployed to execute on one or more computers, the computers being located in one place, or distributed to multiple locations and interconnected through a communication network. The processes and logic flows described in this specification may be performed by one or more programmable computers that execute one or more computer programs to run functions by computing input data and generating output.

[0036] The number of devices and processing scales described here are used to simplify the description of the present invention. Applications, modifications and variations of the present invention will be obvious to those skilled in the art.

[0037] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the details shown and described herein.

Claims

1. An intelligent driving obstacle avoidance processing method, characterized in that: include: Step 1: determine whether the obstacle in front of the intelligent driving vehicle is within the planned driving path of the intelligent driving vehicle. If not, continue to drive according to the planned driving path. If yes, proceed to step 2. Step 2: Detect the relative distance between the intelligent driving vehicle and the obstacle in real time. Once it is less than the set braking safety distance, detect the lane line to determine whether the lane can be changed. If it is, execute the lane change obstacle avoidance decision, and further calculate the variable lane vehicle distance. If it is greater than the set lane change safety distance, execute the lane change action; otherwise, execute the deceleration obstacle avoidance decision.

2. The intelligent driving obstacle avoidance processing method according to claim 1, characterized in that: The set braking safety distance is twice the minimum braking distance. The minimum braking distance is calculated as follows: S= Where S represents the minimum braking distance, Represents the speed of the intelligent driving vehicle, Represents the friction coefficient between the intelligent driving vehicle and the road surface, represents the acceleration due to gravity, represents the correction factor, obtained through empirical data, Represents the actual load weight of the smart driving vehicle, Represents the unladen weight of the smart driving vehicle, Represents the response time of the intelligent driving vehicle braking command system, Represents the error factor, obtained through empirical data.

3. The intelligent driving obstacle avoidance processing method according to claim 1, characterized in that: The set lane change safety distance is twice the minimum lane change distance. The calculation of the minimum lane change distance is achieved by constructing a BP neural network. Specifically, a data set is constructed, including the longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the target lane risk vehicle, the lane change time, the actual load weight of the intelligent driving vehicle, the empty weight of the intelligent driving vehicle, and the lane change distance of the intelligent driving vehicle. The data set is divided into a training set, a validation set, and a test set to construct a BP neural network. The longitudinal acceleration and lateral acceleration of the intelligent driving vehicle, the longitudinal acceleration and lateral acceleration of the target lane risk vehicle, the lane change time, the actual load weight of the intelligent driving vehicle, and the empty weight of the intelligent driving vehicle are used as the input layer of the BP neural network, and the lane change distance of the intelligent driving vehicle is used as the output layer of the BP neural network. The number of hidden layers is set to 10. The advantages and disadvantages of the BP neural network are judged according to the correlation and error between the predicted value and the measured value, and the minimum lane change distance is predicted based on the BP neural network.

4. The intelligent driving obstacle avoidance processing method according to claim 1, characterized in that: The obstacle directly in front of the smart driving vehicle is judged by radar and camera. The radar is used to detect the relative distance between the smart driving vehicle and the obstacle directly in front of it, and also to detect the relative distance between the smart driving vehicle and the target lane risk vehicles on the left and right lanes of its driving direction. The camera is used to detect lane lines to determine whether lanes can be changed, and also to photograph the driving environment of the smart driving vehicle.

5. The intelligent driving obstacle avoidance processing method as described in claim 4 is characterized in that the frequency of the radar acquiring distance information is 10 Hz, and once it is less than the set braking safety distance, the frequency of the radar acquiring distance information is 100 Hz.

6. The intelligent driving obstacle avoidance processing method according to claim 4, characterized in that: The speed of the obstacle is detected in real time, and it is determined whether the obstacle is a static obstacle or a dynamic obstacle. If it is a static obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time. If it is a dynamic obstacle, the relative distance between the intelligent driving vehicle and the obstacle is detected in real time. The speed of the intelligent driving vehicle is detected in real time, and the relative distance between the intelligent driving vehicle and the obstacle is calculated in real time based on the speed of the obstacle. The smaller value of the detected relative distance and the calculated relative distance is taken as the relative distance between the intelligent driving vehicle and the obstacle for subsequent steps.

7. An electronic device, characterized in that include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the method described in any one of claims 1 to 6.

8. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method described in any one of claims 1 to 6 is implemented.

9. A computer program product comprising a computer program, characterized in that When the program is executed by a processor, the method described in any one of claims 1 to 6 is implemented.

Images