Intelligent autonomous obstacle avoidance method and system for unmanned driving roads based on behavior perception
By collecting and analyzing weather, vehicle and road data, we build a prediction model of obstacle avoidance behavior, which solves the problem of failure to effectively analyze object behavior trends in driverless technology, improves the efficiency of obstacle avoidance decisions, and achieves more accurate obstacle avoidance behaviors implement.
Patent Information
- Application Number
- CN202410751504.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-12
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-06-12
AI Technical Summary
The existing unmanned driving technology has failed to effectively analyze the parameters such as the motion state, speed and acceleration of objects to infer the behavioral trend of objects, resulting in low efficiency in obstacle avoidance decisions.
Adopt the intelligent autonomous obstacle avoidance method of unmanned driving roads based on behavior perception, and calculate the vehicle obstacle avoidance efficiency value and road obstacle avoidance efficiency value by collecting weather data, vehicle data and road data, build an obstacle avoidance behavior prediction model, predict and execute avoidance Obstacle behavior.
It improves the work efficiency of driverless cars in making obstacle avoidance decisions, can predict and execute obstacle avoidance behavior more accurately, and improves intelligent obstacle avoidance capabilities.
Smart Images

Figure CN118323196B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent driving technology, and in particular to an intelligent autonomous obstacle avoidance method and system for unmanned driving roads based on behavior perception. Background Art
[0002] With the rapid development of artificial intelligence, computer vision, sensor technology and other fields, driverless technology has also developed rapidly. Driverless cars can perceive the surrounding environment in real time by carrying various sensors, such as lidar, cameras, ultrasonic sensors, etc., which provides a basis for intelligent autonomous obstacle avoidance.
[0003] For example, a Chinese patent with the authorization announcement number CN113031583B discloses a structured road obstacle avoidance method, which includes: obtaining the vehicle posture in real-time driving, the obstacle state information around the driving environment, and the reference path information, based on which, the visible points of the obstacle in front of the current vehicle are sampled in state space to determine the posture of each sampling point; taking the posture of each sampling point as the terminal state and the current vehicle posture as the initial state, the current vehicle posture is connected with the posture of each sampling point respectively to obtain the motion primitive of the current vehicle relative to each sampling point; constructing a structured road cost function including path smoothness, collision risk possibility, and reference line path followability, and using the cost function, selecting the optimal motion primitive and determining it as the obstacle avoidance path for the current obstacle. This invention realizes the autonomous obstacle avoidance of unmanned mining trucks during structured road driving, improves the intelligence level of mines, and realizes autonomous and safe operation of mining trucks.
[0004] For example, a Chinese patent application with publication number CN116360422A discloses a control method for an automatic cruising intelligent obstacle avoidance car, which places the intelligent car on the driving road of a traffic sandbox and collects real-time road condition images; the method includes the following steps: S1, analyzing the road condition image to identify the information of objects on the road; S2, estimating the distance between the intelligent car and each object with height in combination with the real height of each object stored in advance; S3, if there is a traffic light or pedestrian in a red light state among the identified objects and the distance from the car is less than the corresponding distance threshold, then go to S4, otherwise go to S5; S4, execute the preset waiting strategy; S5, if there is an obstacle among the identified objects and the distance from the car is less than the preset safe avoidance distance, then execute the preset obstacle avoidance driving strategy; otherwise go to S6; S6, execute the preset autonomous cruising driving. This method can improve the intelligence level of the intelligent car and its ability to respond to traffic events.
[0005] The above patents all have the problem raised by this background technology: they do not infer the behavior trend of the object by analyzing the object's motion state, speed, acceleration and other parameters; to solve these problems, the present invention proposes an unmanned driving road intelligent autonomous obstacle avoidance method and system based on behavior perception. Summary of the invention
[0006] In view of the shortcomings of the prior art, the main purpose of the present invention is to provide an unmanned driving road intelligent autonomous obstacle avoidance method and system based on behavior perception, which can effectively solve the problems in the background technology. The specific technical solutions of the present invention are as follows:
[0007] The method for intelligent autonomous obstacle avoidance on unmanned driving roads based on behavior perception includes the following specific steps:
[0008] S1. Collect weather data, vehicle data, and road data of road sections at m times, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data;
[0009] S2. Substituting weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value;
[0010] S3, substituting the vehicle obstacle avoidance efficiency value and the road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section;
[0011] S4, substituting the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtaining the obstacle avoidance behavior that the unmanned vehicle needs to perform;
[0012] S5. Transmitting the obstacle avoidance behavior to be executed to the vehicle control system, and the vehicle control system executes the obstacle avoidance behavior to be executed.
[0013] Specifically, the tire data includes: tire pressure, tire surface friction coefficient, the surrounding obstacle data includes: stationary obstacle data, moving obstacle data, the stationary obstacle data includes the number of stationary obstacles, and the real-time distance of each stationary obstacle from the vehicle, the moving obstacle data includes the number of moving obstacles, and the real-time moving speed of each moving obstacle, and the real-time distance from the vehicle.
[0014] Specifically, the road section data is the average travel speed of the road section, and the specific calculation steps of the average travel speed of the road section are:
[0015] S11, obtaining driving data of n vehicles before the driverless vehicle enters the road section, the driving data including the time required for the n vehicles to pass through the road section, and obtaining the length of the road section;
[0016] S12: Average travel speed of the road section , where L is the length of the road section, It is the time required for the i-th vehicle to pass through the road section, where i is any item from 1 to n.
[0017] Specifically, the specific calculation steps of the vehicle obstacle avoidance efficiency value calculation strategy in S2 are:
[0018] S21, calculating the road slipperiness value at the jth moment according to the rainfall, snowfall, and temperature of the road section at the jth moment ,in are the rainfall influence coefficient and the snowfall influence coefficient respectively. And they are all greater than or equal to 0; max() means taking the maximum value in the brackets, R is the rainfall at the jth moment, S is the snowfall at the jth moment, and T is the temperature at the jth moment. They are the thresholds for rainfall and snowfall to affect the road surface, respectively;
[0019] S22, calculating the dynamic response capability value of the unmanned vehicle at the jth moment according to the speed of the unmanned vehicle and the acceleration of the unmanned vehicle on the road section at the jth moment ,in are velocity influence coefficient and acceleration influence coefficient respectively. and are both greater than or equal to 0; V is the speed of the unmanned vehicle on the road section at the jth moment, A is the acceleration of the unmanned vehicle on the road section at the jth moment, is the maximum vehicle speed allowed on the road section, is the maximum acceleration of the driverless vehicle;
[0020] S23, calculating the vehicle grip efficiency value at the jth moment according to the tire pressure and tire surface friction coefficient of the unmanned vehicle at the jth moment ,in They are tire pressure influence coefficient and tire friction influence coefficient respectively. And they are all greater than or equal to 0; min() means taking the minimum value in the brackets, P is the tire pressure of the unmanned vehicle at the jth moment, u is the friction coefficient of the tire surface of the unmanned vehicle at the jth moment, To ensure the maximum tire pressure for driverless vehicles, To ensure the minimum tire pressure for driverless vehicles.
[0021] Specifically, the vehicle obstacle avoidance efficiency value calculation strategy also includes:
[0022] Calculate the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment ,in They are the influence coefficient of road slipperiness, the influence coefficient of vehicle dynamic response capability, and the influence coefficient of vehicle grip efficiency. , and both are greater than 0.
[0023] Specifically, S3 includes the following specific steps:
[0024] S31, obtaining the number of stationary obstacles in a circular area with a radius of r centered on the unmanned vehicle at the jth moment, as well as the real-time distance of each stationary obstacle from the vehicle, the number of mobile obstacles, as well as the real-time moving speed of each mobile obstacle, the real-time distance from the vehicle, the average travel speed of the road section, and the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment;
[0025] S32, then the road obstacle avoidance efficiency value of the unmanned vehicle in the road section at the jth moment ,in are the influence coefficient of stationary obstacles and the influence coefficient of moving obstacles respectively. and are both greater than or equal to 0; Lw is the distance that the driverless vehicle has traveled on the road section at the jth moment, is the real-time distance between the qth stationary obstacle and the vehicle, is the number of stationary obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, is the real-time distance and real-time moving speed of the pth moving obstacle from the vehicle, is the number of moving obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, q is 1 to Any item in Any one of .
[0026] Specifically, the S4 specifically includes the following contents:
[0027] Obtain the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments in history, and the corresponding obstacle avoidance behaviors, and establish an obstacle avoidance behavior data set for storing the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments, and the corresponding obstacle avoidance behaviors; the unmanned vehicle autonomous obstacle avoidance analysis strategy predicts the obstacle avoidance behaviors that the unmanned vehicle needs to perform by constructing an obstacle avoidance behavior prediction model, and divides the obstacle avoidance behavior data set into an obstacle avoidance behavior training set and an obstacle avoidance behavior test set; construct a behavior prediction model, use the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the input of the behavior prediction model, use the obstacle avoidance behaviors corresponding to the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the output of the behavior prediction model, train the behavior prediction model, and obtain an initial obstacle avoidance behavior prediction model; verify the initial obstacle avoidance behavior prediction model through the obstacle avoidance behavior test set, and output the initial obstacle avoidance behavior prediction model with an accuracy greater than or equal to a preset model as the obstacle avoidance behavior prediction model.
[0028] Specifically, the unmanned driving road intelligent autonomous obstacle avoidance system based on behavior perception is implemented based on the unmanned driving road intelligent autonomous obstacle avoidance method based on behavior perception, and the system includes the following modules:
[0029] A data collection module, used to collect weather data, vehicle data, and road data of road sections at m moments, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data;
[0030] A vehicle obstacle avoidance efficiency value calculation module is used to substitute weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value;
[0031] A road obstacle avoidance efficiency value calculation module is used to substitute the vehicle obstacle avoidance efficiency value and road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section;
[0032] The obstacle avoidance behavior analysis module is used to substitute the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtain the obstacle avoidance behavior that the unmanned vehicle needs to perform;
[0033] An obstacle avoidance behavior execution module, used for transmitting the obstacle avoidance behavior to be executed to the vehicle control system, so that the vehicle control system executes the obstacle avoidance behavior to be executed;
[0034] The control module is used to control the operation of each module.
[0035] Compared with the prior art, the present invention has the following beneficial effects:
[0036] The present invention collects weather data, vehicle data, and road data, substitutes the weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value; substitutes the vehicle obstacle avoidance efficiency value and road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section; substitutes the road obstacle avoidance efficiency value into the unmanned vehicle autonomous obstacle avoidance analysis strategy for analysis, and obtains the obstacle avoidance behavior that the unmanned vehicle needs to perform; transmits the obstacle avoidance behavior that needs to be performed to the vehicle control system, and the vehicle control system performs the obstacle avoidance behavior that needs to be performed. The present invention can improve the working efficiency of unmanned vehicles when making obstacle avoidance decisions. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 The flowchart of the intelligent autonomous obstacle avoidance method for unmanned driving on a road based on behavior perception of the present invention;
[0038] Figure 2 The module diagram of the unmanned driving road intelligent autonomous obstacle avoidance system based on behavior perception of the present invention. DETAILED DESCRIPTION
[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.
[0040] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0041] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
[0042] Example 1
[0043] This embodiment provides an intelligent autonomous obstacle avoidance method for unmanned driving on a road based on behavior perception. The specific scheme is as follows: Figure 1 As shown, the intelligent autonomous obstacle avoidance method for unmanned driving roads based on behavior perception includes the following specific steps:
[0044] S1. Collect weather data, vehicle data, and road data of road sections at m times, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data;
[0045] S2. Substituting weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value;
[0046] S3, substituting the vehicle obstacle avoidance efficiency value and the road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section;
[0047] S4, substituting the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtaining the obstacle avoidance behavior that the unmanned vehicle needs to perform;
[0048] S5. Transmitting the obstacle avoidance behavior to be executed to the vehicle control system, and the vehicle control system executes the obstacle avoidance behavior to be executed.
[0049] In this embodiment, the tire data includes: tire pressure, tire surface friction coefficient, and the surrounding obstacle data includes: stationary obstacle data and moving obstacle data. The stationary obstacle data includes the number of stationary obstacles and the real-time distance of each stationary obstacle from the vehicle. The moving obstacle data includes the number of moving obstacles and the real-time moving speed of each moving obstacle and the real-time distance from the vehicle.
[0050] In this embodiment, the surrounding obstacles may include: road signs, traffic lights, shoulders, bridges, tunnel entrances, buildings along the road or at intersections, such as houses, shops, office buildings, trees on both sides of the road, bushes, lawns, electric poles, bus stops, mailboxes, fire hydrants, vehicles parked on the roadside or in parking lots, roadblocks, construction fences, discarded objects, cars, trucks, motorcycles, bicycles driving on the road, pedestrians crossing the road or walking on the road, as well as non-motorized vehicles and animals appearing on the road;
[0051] In this embodiment, the surrounding obstacle data is acquired through various sensors built into the unmanned vehicle, including laser radar, camera and ultrasonic sensor.
[0052] In this embodiment, the road section data is the average travel speed of the road section, and the specific calculation steps of the average travel speed of the road section are:
[0053] S11, obtaining driving data of n vehicles before the driverless vehicle enters the road section, the driving data including the time required for the n vehicles to pass through the road section, and obtaining the length of the road section;
[0054] S12: Average travel speed of the road section , where L is the length of the road section, It is the time required for the i-th vehicle to pass through the road section, where i is any item from 1 to n.
[0055] In this embodiment, the specific calculation steps of the vehicle obstacle avoidance efficiency value calculation strategy in S2 are:
[0056] S21, calculating the road slipperiness value at the jth moment according to the rainfall, snowfall, and temperature of the road section at the jth moment ,in are respectively the rainfall influence coefficient and the snowfall influence coefficient, which are set by technicians in this field according to actual conditions. And they are all greater than or equal to 0; max() means taking the maximum value in the brackets, R is the rainfall at the jth moment, S is the snowfall at the jth moment, and T is the temperature at the jth moment. They are the thresholds for rainfall and snowfall to affect the road surface, respectively; The units of are consistent with those of T. It is determined by technical personnel in this field through a large number of tests to determine whether the effects of rainfall and snowfall on the slipperiness of the road surface are within a safe range.
[0057] S22, calculating the dynamic response capability value of the unmanned vehicle at the jth moment according to the speed of the unmanned vehicle and the acceleration of the unmanned vehicle on the road section at the jth moment ,in They are speed influence coefficient and acceleration influence coefficient respectively, which are set by technicians in this field according to actual conditions. And they are both greater than or equal to 0, V is the speed of the unmanned vehicle on the road section at the jth moment, A is the acceleration of the unmanned vehicle on the road section at the jth moment, is the maximum vehicle speed allowed on the road section, is the maximum acceleration of the driverless vehicle;
[0058] S23, calculating the vehicle grip efficiency value at the jth moment according to the tire pressure and tire surface friction coefficient of the unmanned vehicle at the jth moment ,in They are tire pressure influence coefficient and tire friction influence coefficient, which are set by technicians in this field according to actual conditions. And they are all greater than or equal to 0; min() means taking the minimum value in the brackets, P is the tire pressure of the unmanned vehicle at the jth moment, u is the friction coefficient of the tire surface of the unmanned vehicle at the jth moment, To ensure the maximum tire pressure for driverless vehicles, To ensure the minimum tire pressure for driverless vehicles.
[0059] In this embodiment, the vehicle obstacle avoidance efficiency value calculation strategy also includes:
[0060] Calculate the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment ,in They are the influence coefficient of road slipperiness, the influence coefficient of vehicle dynamic response capability, and the influence coefficient of vehicle grip efficiency, which are set by technicians in this field according to actual conditions. , and both are greater than 0.
[0061] In this embodiment, S3 includes the following specific steps:
[0062] S31, obtaining the number of stationary obstacles in a circular area with a radius of r centered on the unmanned vehicle at the jth moment, as well as the real-time distance of each stationary obstacle from the vehicle, the number of mobile obstacles, as well as the real-time moving speed of each mobile obstacle, the real-time distance from the vehicle, the average travel speed of the road section, and the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment;
[0063] S32, then the road obstacle avoidance efficiency value of the unmanned vehicle in the road section at the jth moment ,in are the stationary obstacle influence coefficient and the moving obstacle influence coefficient, respectively, and are set by technicians in this field according to actual conditions. and are both greater than or equal to 0; Lw is the distance that the driverless vehicle has traveled on the road section at the jth moment, is the real-time distance between the qth stationary obstacle and the vehicle, is the number of stationary obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, is the real-time distance and real-time moving speed of the pth moving obstacle from the vehicle, is the number of moving obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, q is 1 to Any item in Any one of .
[0064] In this embodiment, S4 specifically includes the following contents:
[0065] Obtain the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments in history, and the corresponding obstacle avoidance behaviors, and establish an obstacle avoidance behavior data set for storing the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments, and the corresponding obstacle avoidance behaviors; the unmanned vehicle autonomous obstacle avoidance analysis strategy predicts the obstacle avoidance behaviors that the unmanned vehicle needs to perform by constructing an obstacle avoidance behavior prediction model, and divides the obstacle avoidance behavior data set into an obstacle avoidance behavior training set and an obstacle avoidance behavior test set; construct a behavior prediction model, use the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the input of the behavior prediction model, use the obstacle avoidance behaviors corresponding to the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the output of the behavior prediction model, train the behavior prediction model, and obtain an initial obstacle avoidance behavior prediction model; verify the initial obstacle avoidance behavior prediction model through the obstacle avoidance behavior test set, and output the initial obstacle avoidance behavior prediction model with an accuracy greater than or equal to a preset model as the obstacle avoidance behavior prediction model.
[0066] In this embodiment, the obstacle avoidance behavior is the obstacle avoidance behavior monitored by the vehicle system at m moments, and the obstacle avoidance behavior includes: turning, accelerating, decelerating, emergency braking, changing lanes, and maintaining the current driving state. The obstacle avoidance behavior prediction model is any one of a recurrent neural network model, a convolutional neural network model, and a deep reinforcement learning model.
[0067] Example 2
[0068] This embodiment provides an unmanned driving road intelligent autonomous obstacle avoidance system based on behavior perception, and the specific scheme is as follows: Figure 2 As shown, the unmanned driving road intelligent autonomous obstacle avoidance system based on behavior perception is implemented based on the unmanned driving road intelligent autonomous obstacle avoidance method based on behavior perception described in Example 1, and the system includes the following modules:
[0069] A data collection module, used to collect weather data, vehicle data, and road data of road sections at m moments, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data;
[0070] A vehicle obstacle avoidance efficiency value calculation module is used to substitute weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value;
[0071] A road obstacle avoidance efficiency value calculation module is used to substitute the vehicle obstacle avoidance efficiency value and road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section;
[0072] The obstacle avoidance behavior analysis module is used to substitute the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtain the obstacle avoidance behavior that the unmanned vehicle needs to perform;
[0073] An obstacle avoidance behavior execution module, used for transmitting the obstacle avoidance behavior to be executed to the vehicle control system, so that the vehicle control system executes the obstacle avoidance behavior to be executed;
[0074] The control module is used to control the operation of each module.
[0075] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. An intelligent autonomous obstacle avoidance method for unmanned driving based on behavior perception, characterized by: The method comprises the following specific steps: S1. Collect weather data, vehicle data, and road data of road sections at m times, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data; S2. Substituting weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value; S3, substituting the vehicle obstacle avoidance efficiency value and the road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section; S4, substituting the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtaining the obstacle avoidance behavior that the unmanned vehicle needs to perform; S5, transmitting the obstacle avoidance behavior to be executed to the vehicle control system, and the vehicle control system executes the obstacle avoidance behavior to be executed; The road section data is the average travel speed of the road section, and the specific calculation steps of the average travel speed of the road section are: S11, obtaining driving data of n vehicles before the driverless vehicle enters the road section, the driving data including the time required for the n vehicles to pass through the road section, and obtaining the length of the road section; S12: Average travel speed of the road section , where L is the length of the road section, is the time required for the i-th vehicle to pass through the road section, where i is any one from 1 to n; The specific calculation steps of the vehicle obstacle avoidance efficiency value calculation strategy in S2 are: S21, calculating the road slipperiness value at the jth moment according to the rainfall, snowfall, and temperature of the road section at the jth moment ,in are the rainfall influence coefficient and the snowfall influence coefficient respectively. And they are all greater than or equal to 0; max() means taking the maximum value in the brackets, R is the rainfall at the jth moment, S is the snowfall at the jth moment, and T is the temperature at the jth moment. They are the thresholds for rainfall and snowfall to affect the road surface, respectively; S22, calculating the dynamic response capability value of the unmanned vehicle at the jth moment according to the speed of the unmanned vehicle and the acceleration of the unmanned vehicle on the road section at the jth moment ,in are velocity influence coefficient and acceleration influence coefficient respectively. and are both greater than or equal to 0; V is the speed of the unmanned vehicle on the road section at the jth moment, A is the acceleration of the unmanned vehicle on the road section at the jth moment, is the maximum vehicle speed allowed on the road section, is the maximum acceleration of the driverless vehicle; S23, calculating the vehicle grip efficiency value at the jth moment according to the tire pressure and tire surface friction coefficient of the unmanned vehicle at the jth moment ,in They are tire pressure influence coefficient and tire friction influence coefficient respectively. And they are all greater than or equal to 0; min() means taking the minimum value in the brackets, P is the tire pressure of the unmanned vehicle at the jth moment, u is the friction coefficient of the tire surface of the unmanned vehicle at the jth moment, To ensure the maximum tire pressure for driverless vehicles, To ensure the minimum tire pressure for driverless vehicles; The vehicle obstacle avoidance efficiency value calculation strategy also includes: Calculate the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment ,in They are the influence coefficient of road slipperiness, the influence coefficient of vehicle dynamic response capability, and the influence coefficient of vehicle grip efficiency. , and are all greater than 0; S3 includes the following specific steps: S31, obtaining the number of stationary obstacles in a circular area with a radius of r centered on the unmanned vehicle at the jth moment, as well as the real-time distance of each stationary obstacle from the vehicle, the number of mobile obstacles, as well as the real-time moving speed of each mobile obstacle, the real-time distance from the vehicle, the average travel speed of the road section, and the obstacle avoidance efficiency value of the unmanned vehicle at the jth moment; S32, then the road obstacle avoidance efficiency value of the unmanned vehicle in the road section at the jth moment ,in are the influence coefficient of stationary obstacles and the influence coefficient of moving obstacles respectively. and are both greater than or equal to 0; Lw is the distance that the driverless vehicle has traveled on the road section at the jth moment, is the real-time distance between the qth stationary obstacle and the vehicle, is the number of stationary obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, is the real-time distance and real-time moving speed of the pth moving obstacle from the vehicle, is the number of moving obstacles in the circular area with a radius of r centered on the driverless vehicle at the jth moment, q is 1 to Any item in Any one of .
2. The method for intelligent autonomous obstacle avoidance on unmanned roads based on behavior perception according to claim 1 is characterized by: The tire data includes: tire pressure, tire surface friction coefficient, and the surrounding obstacle data includes: stationary obstacle data and moving obstacle data. The stationary obstacle data includes the number of stationary obstacles and the real-time distance of each stationary obstacle from the vehicle. The moving obstacle data includes the number of moving obstacles and the real-time moving speed of each moving obstacle and the real-time distance from the vehicle.
3. The method for intelligent autonomous obstacle avoidance on unmanned roads based on behavior perception according to claim 2 is characterized by: The S4 specifically includes the following contents: Obtain the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments in history, and the corresponding obstacle avoidance behaviors, and establish an obstacle avoidance behavior data set for storing the road obstacle avoidance efficiency values of unmanned vehicles on road sections at m moments, and the corresponding obstacle avoidance behaviors; the unmanned vehicle autonomous obstacle avoidance analysis strategy predicts the obstacle avoidance behaviors that the unmanned vehicle needs to perform by constructing an obstacle avoidance behavior prediction model, and divides the obstacle avoidance behavior data set into an obstacle avoidance behavior training set and an obstacle avoidance behavior test set; construct a behavior prediction model, use the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the input of the behavior prediction model, use the obstacle avoidance behaviors corresponding to the road obstacle avoidance efficiency values at all moments in the obstacle avoidance behavior training set as the output of the behavior prediction model, train the behavior prediction model, and obtain an initial obstacle avoidance behavior prediction model; verify the initial obstacle avoidance behavior prediction model through the obstacle avoidance behavior test set, and output the initial obstacle avoidance behavior prediction model with an accuracy greater than or equal to a preset model as the obstacle avoidance behavior prediction model.
4. An unmanned driving road intelligent autonomous obstacle avoidance system based on behavior perception, which is implemented based on an unmanned driving road intelligent autonomous obstacle avoidance method based on behavior perception as claimed in any one of claims 1 to 3, characterized in that: The system includes the following modules: A data collection module, used to collect weather data, vehicle data, and road data of road sections at m moments, wherein the weather data includes rainfall, snowfall, and temperature; the vehicle data includes unmanned vehicle speed, unmanned vehicle acceleration, and tire data; and the road data includes surrounding obstacle data and road section data; A vehicle obstacle avoidance efficiency value calculation module is used to substitute weather data and vehicle data into the vehicle obstacle avoidance efficiency value calculation strategy to calculate the vehicle obstacle avoidance efficiency value; A road obstacle avoidance efficiency value calculation module is used to substitute the vehicle obstacle avoidance efficiency value and road data into the road obstacle avoidance efficiency value calculation strategy to calculate the road obstacle avoidance efficiency value of the unmanned vehicle in the road section; The obstacle avoidance behavior analysis module is used to substitute the road obstacle avoidance efficiency value into the autonomous obstacle avoidance analysis strategy of the unmanned vehicle for analysis, and obtain the obstacle avoidance behavior that the unmanned vehicle needs to perform; An obstacle avoidance behavior execution module, used for transmitting the obstacle avoidance behavior to be executed to the vehicle control system, so that the vehicle control system executes the obstacle avoidance behavior to be executed; The control module is used to control the operation of each module.
