Intelligent networked automobile decision planning module simulation test system and method thereof

Abstract

The invention discloses an intelligent networked automobile decision planning module simulation test system and a method thereof. The system comprises a driving simulator, RSU simulation equipment, an unmanned simulation automobile, an intelligent driving database and a data processing unit. The driving simulator comprises a parameter setting unit, a scene calculation unit, a real-time scene display unit, a GPS simulator and a vehicle state transmission unit. The RSU simulation equipment comprises a host and a radio frequency unit; the unmanned simulation automobile comprises an on-board unit (OBU), a vehicle data memory, a positioning sensor, a to-be-tested decision planning module, a path parameter setting unit and an Ethernet interface; the intelligent driving database comprises a high-precision map database, a traffic rule database and a scene database; and the data processing unit comprises a data memory and a data analysis system. The method saves site facilities, does not need to build complex instruments and equipment, and is low in cost, flexible in application and high in efficiency.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a simulation test system and method for a decision-making planning module of an intelligent networked vehicle. Background Art

[0002] With the rapid development of society and economy, cars have become an increasingly important means of transportation. However, the increase in the number of cars and the frequency of human car travel has brought tremendous pressure to traffic safety, and road traffic accidents have occurred frequently. Autonomous driving technology is an effective way to improve driving safety and prevent and reduce road traffic accidents. By equipping cars with advanced on-board sensors, controllers, actuators and other devices, and based on the fusion environment perception technology of multiple sensors such as images, radars, satellite differential positioning, and inertial navigation, the vehicle can automatically avoid obstacles, improve the safety of vehicle driving, protect the driver's life and property safety, and improve traffic efficiency while meeting the growing traffic demand.

[0003] Intelligent networking is an effective way to achieve autonomous driving. Using advanced wireless communications and new generation Internet technologies, we build a connected vehicle system that connects vehicles, pedestrians, roadside equipment, and service cloud platforms. Through real-time dynamic information interaction between vehicles, people, roads, and clouds, we can effectively guide and supervise vehicles. Roadside equipment uses laser radar, millimeter wave radar, machine vision terminals, and ultrasonic radar to collect data on roads, pedestrians, obstacles, etc. in real time, so that smart cars can obtain information on vulnerable traffic participants in the visual blind spot in advance, and make timely warnings and reasonable decisions on sudden traffic conditions, effectively avoiding traffic accidents and improving traffic efficiency.

[0004] The decision-making and planning module is a key part of intelligent connected vehicles. It is mainly used to carry out global path planning, behavioral decision-making, and local path planning to ensure that the vehicle can safely bypass obstacles or perform obstacle avoidance actions, so that autonomous driving vehicles can drive safely and efficiently in complex and changeable traffic environments.

[0005] Carrying out performance testing and functional verification of the decision-making and planning modules of intelligent connected vehicles in complex traffic scenarios is a necessary task for promoting the construction of intelligent transportation systems. Compared with real road tests, using computer simulation technology to establish a simulation test system in the laboratory to carry out decision-making and planning simulation and testing has the advantages of economy, safety, repeatability, controllability, and scalability. At present, the patents disclosed in this field mainly focus on the decision-making and planning architecture and implementation methods of intelligent connected vehicles. Beijing Sankuai Online Technology Co., Ltd. proposed an automatic driving decision-making and planning method in patent CN202211303615.2, which obtains relevant information of the obstacle vehicle and judges its driving intention to avoid collision. Tsinghua University proposed an automatic driving decision-making and planning architecture based on vehicle-road-cloud integration in patent CN202410340531.9 to solve the problem that the vehicle-side computing of the current vehicle-road-cloud integrated architecture is heavily dependent on cloud computing. Beijing Institute of Technology proposed a compliance safety decision-making planning system and method for autonomous driving vehicles in patent CN202311090175.1, which specifies the specific composition and functions of the perception module, navigation module, compliance safety decision-making planning module and control module, and provides a decision-making planning implementation method. The above patent has not yet provided a test method for the decision-making planning performance of intelligent connected vehicles. Summary of the invention

[0006] In response to the problems existing in the prior art, the present invention provides a simulation test system and method for a decision-making and planning module of an intelligent connected vehicle, which saves site facilities, does not require the construction of complex instruments and equipment, has low cost, flexible application, and high efficiency.

[0007] The purpose of the present invention is achieved through the following technical solutions.

[0008] A simulation test system for a decision-making planning module of an intelligent networked vehicle comprises: a driving simulator, an RSU simulation device, an unmanned driving simulation vehicle, an intelligent driving database, and a data processing unit.

[0009] The driving simulator includes a parameter setting unit, a scene calculation unit, a real-time scene display unit, a GPS simulator, and a vehicle status transmission unit; the RSU simulation device includes a host and a radio frequency unit; the unmanned driving simulation car includes an on-board unit OBU, a vehicle data storage device, a positioning sensor, a decision-making planning module to be tested, a path parameter setting unit, and an Ethernet interface; the intelligent driving database includes a high-precision map library, a traffic rules database, and a scene database; the data processing unit includes a data storage device and a data analysis system.

[0010] The driving simulator is used to configure the initial state of the car and the static traffic environment of the driving area according to the parameters set by the user, calculate the real-time driving state of the vehicle and the dynamic traffic environment, and display the simulated driving scene in real time. The parameter setting unit is used for the user to set the simulated car driving area, the simulated car initial position, the simulated car model and size. The scene calculation unit generates the static traffic environment of the driving area according to the initial parameters set by the parameter setting unit and the intelligent driving database, initializes the state of the traffic lights at the intersection, initializes the dynamic electronic sign signs, and initializes the traffic state of the road section. During the simulation test, based on the traffic scene at the last moment, the position, speed, acceleration, and direction of the simulated car are dynamically calculated and updated in combination with the current state of the car, the static traffic environment, and the behavior decision results. A new simulated traffic scene is generated. The updated information is the current position, speed, acceleration, direction, and current simulated traffic scene of the simulated vehicle. The scene calculation unit sends the current position of the vehicle to the GPS simulator and sends the current traffic scene data to the real-time scene display unit. The road, environment, and roadside equipment data in the current traffic scene are sent to the RSU simulation device host, the updated position of the vehicle is sent to the GPS simulator, and the updated speed and direction information of the vehicle is sent to the vehicle state transmission unit. The vehicle state transmission unit transmits the vehicle size, speed, and direction information to the vehicle data storage device through the Ethernet interface of the unmanned driving simulation vehicle in an Ethernet wired manner, and the vehicle data storage device updates the vehicle real-time size, speed, and direction information therein. The real-time scene display unit displays the virtual scene of the road ahead on the display screen from the observation angle of the simulation vehicle based on the real-time scene data generated by the scene calculation unit, and updates it in real time.

[0011] The intelligent driving database provides the required high-precision map data, traffic scene data, and traffic rules data for the driving simulator and the unmanned driving simulation car decision-making and planning module. The scenario simulation database includes vehicle-road collaborative scenarios, which are used to simulate green wave speed guidance, front congestion reminders, road hazard reminders, speed limit sign reminders, construction sign reminders, red light running warnings, blind spot pedestrian crossing warnings, ramp vehicle merging warnings and other scenarios. The high-precision map library is static high-precision map data with lane-level accuracy, and the map covers an area of ​​no less than 100 square kilometers. The traffic rules database contains traffic regulations that motor vehicles, non-motor vehicles, and pedestrians should abide by when traveling on the road as stipulated by the national traffic management department.

[0012] The RSU simulation device is used to communicate with the OBU of the unmanned driving simulation car. The host generates real-time data of the roadside unit in the simulation scene based on the roadside equipment data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit. The radio frequency unit broadcasts RSU information in a wireless transmission manner and communicates with the on-board terminal OBU of the unmanned driving simulation car.

[0013] The unmanned driving simulation car is a simulation system with functions such as sensing traffic environment information, vehicle status, path planning and behavior decision-making. The OBU is an on-board unit device, which receives the road traffic environment information sent by the RSU simulator in a wireless transmission mode and saves this information in the vehicle data storage. The positioning sensor receives the current position information of the vehicle sent by the GPS simulator and updates the vehicle position information in the vehicle data storage at the same time. The vehicle data storage stores and updates the road traffic environment information, the current position, speed, driving direction and other information of the vehicle transmitted by the OBU in real time. The decision-making planning module to be tested is a decision-making planning module for testing performance. According to the current environment information and vehicle status, path planning and behavior decision-making are carried out, and the drive control, steering control, braking control and other signals required for the next operation of the vehicle are output, and these information are transmitted to the scene calculation unit of the driving simulator in a wired manner through the Ethernet interface. The path parameter setting unit is an interface for receiving user input parameters before the unmanned driving car carries out the decision-making planning test. The input parameters include parameters such as the end point, the must-pass point on the way, and priority conditions such as time, energy consumption, and distance.

[0014] The data storage device stores user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent by the GPS simulator to the positioning sensor, the length, width, height, position, speed, acceleration, direction and other data of the unmanned driving simulation car calculated by the vehicle state transmission unit and sent to the vehicle data storage device, the RSU information broadcast by the RSU simulation device, and the drive control, steering control and brake control signals sent by the decision-making and planning module to be tested to the scene calculation unit. The data analysis system is used to analyze the above stored data and evaluate the performance of the decision-making and planning module to be tested.

[0015] A simulation test method for a decision-making planning module of an intelligent connected vehicle comprises the following steps:

[0016] 1) The user selects the driving area, initial position, model and size of the simulated car through the parameter setting unit of the driving simulator, selects weather parameters such as sunny, rainy or haze, and sets temporary construction sections and prohibited traffic conditions in the driving area as needed. The user selects the car's destination through the unmanned driving simulation car path parameter setting unit, enters the waypoints in sequence, and selects one condition from the priority conditions such as shortest time, least energy consumption, shortest distance, and least traffic lights as the priority basis for path planning.

[0017] 2) The scene calculation unit reads the model data in the scene simulation database, high-precision map library, and traffic rule library in the intelligent driving database according to the parameters set by the driving simulator parameter setting unit, and configures the parameters of the road, traffic environment, pedestrians, other vehicles, etc. in the driving area through calculation, initializes the parameters of the traffic facilities contained in the area, and generates a virtual scene. At the same time, the corresponding roadside unit RSU status data is configured, and the length, width, height, and initial position of the simulated car are configured. The current speed and acceleration of the car are 0, and the initial value of the direction is 0.

[0018] 3) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's length, width, height, speed, acceleration, and direction data to the vehicle state transmission unit.

[0019] 4) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the length, width, height, speed, acceleration, and direction data of the simulated car to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores it in the vehicle data storage. The GPS simulator sends the location data of the simulated car to the positioning sensor via wireless transmission and stores it in the vehicle data storage.

[0020] 5) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit. The RSU simulation device and the OBU of the simulated car exchange information. The radio frequency unit broadcasts the RSU status data by wireless transmission. After receiving it, the OBU stores the roadside unit status data in the vehicle data storage.

[0021] 6) After the user confirms the parameter settings, the decision-making planning module to be tested carries out global path planning. The decision-making planning module to be tested reads the data such as the car driving area, road environment, traffic facilities, initial position and size of the car in the vehicle data storage according to the user input parameters, and then combines the high-precision map library and traffic rules database in the intelligent driving database to carry out global path planning, and obtains a global path that meets the optimal conditions from the starting point, through the waypoints in sequence, and finally to the end point.

[0022] 7) If the current simulated car is in the initial stationary state, the decision-making planning module to be tested makes a decision to start the car. At this time, the speed and acceleration of the simulated car are both zero, and the position is the starting point. Otherwise, the decision-making planning module to be tested will make behavioral decisions based on the road environment, traffic facilities, traffic participant parameters, and car position and size data in the current scene stored in the vehicle data storage, combined with the high-precision map library and traffic rules database in the intelligent driving database, determine the driving behavior that the simulated car should take next, and generate decision instructions.

[0023] 8) Based on the global path, according to the current speed, acceleration, position, direction, and size of the car, combined with the current traffic facilities and road environment parameters, the decision-making planning module to be tested carries out local path planning according to the decision-making instructions, calculates the driving trajectory and trajectory points of the unmanned car in the short future, determines the driving speed, acceleration, and direction of each trajectory point, and calculates and generates the driving control, steering control, braking control and other signals of the car. Afterwards, the decision-making planning module to be tested sends the driving control, steering control, braking control and other signal instructions, the vehicle's driving speed, acceleration, direction, position and other information at the next moment to the scene calculation unit through the Ethernet interface and Ethernet wired transmission.

[0024] 9) After receiving the instruction sent by the decision-making and planning module of the vehicle to be tested, the scene calculation unit updates the speed, acceleration, direction and position of the vehicle. Based on the traffic scene at the last moment, according to the model data in the scene simulation database, high-precision map library and traffic rule library in the intelligent driving database, and updates the current parameters such as roads, traffic environment, pedestrians, other vehicles, etc., updates the virtual scene, and configures the corresponding roadside unit RSU status data.

[0025] 10) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's speed, acceleration, and direction data to the vehicle state transmission unit.

[0026] 11) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the simulated car speed, acceleration, and direction data to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage. The GPS simulator sends the simulated car's location data to the positioning sensor via wireless transmission and stores them in the vehicle data storage. The vehicle data storage updates the simulated car's current speed, acceleration, direction, and location data.

[0027] 12) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit. The RSU simulation device sends the RSU status data to the OBU through the radio frequency unit, and the vehicle data storage device updates the roadside unit status data therein.

[0028] 13) The scene calculation unit determines whether the current simulated car collides with the static or dynamic obstacle. If a collision occurs, execute step 15). Otherwise, execute the next step.

[0029] 14) The scene calculation unit determines whether the current simulated car has reached the end point. If it has reached the end point, execute step 16). Otherwise, return to step 7).

[0030] 15) When simulating a car collision, the behavior decision or local path planning performance of the decision-making planning module to be tested was defective and failed the test.

[0031] 16) The data storage of the data processing unit stores the user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent by the GPS simulator to the positioning sensor, the length, width, height, position, speed, acceleration, direction and other data of the simulated car sent by the vehicle state transmission unit to the vehicle data storage, the RSU real-time status information broadcast by the RSU simulation device, and the drive control, steering control and brake control signals sent by the decision-making and planning module to be tested to the scene calculation unit. The data analysis system analyzes the above stored data to evaluate the performance of the decision-making and planning module to be tested.

[0032] Compared with the prior art, the advantages of the present invention are:

[0033] (1) The present invention only needs to carry out intelligent connected vehicle decision-making and planning performance testing in the laboratory, does not rely on actual test roads, does not require the entire actual unmanned vehicle, and saves site facilities.

[0034] (2) There is no need to build complex instruments and equipment, and the behavioral decision-making and path planning capabilities of the decision-making and planning modules to be tested can be tested more comprehensively at a low cost.

[0035] (3) The present invention sets weather parameters such as sunny, rainy or haze, sets temporary construction sections and prohibited traffic conditions within the driving area, and conducts tests in complex scenarios, which is flexible in application. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a structural diagram of the simulation test system of the intelligent networked vehicle decision-making planning module of the present invention.

[0037] Figure 2 It is a simulation test flow chart of the intelligent networked vehicle decision-making planning module of the present invention.

[0038] Figure 3 This is a schematic diagram of the test scenario of Example 1.

[0039] Figure 4 This is a schematic diagram of the test scenario of Example 2. DETAILED DESCRIPTION

[0040] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0041] The decision-making and planning module of intelligent networked vehicles has three key functions: global path planning, behavior decision-making, and local path planning. The purpose of the present invention is to propose a test system and method for the decision-making and planning module of intelligent networked vehicles, generate test scenarios in the laboratory using a simulation system, generate simulated traffic information of roads, environments, pedestrians, and vehicles, and conduct tests on the performance of the decision-making and planning module of intelligent networked vehicles.

[0042] like Figure 1 As shown, a simulation test system for a decision-making planning module of an intelligent connected vehicle includes: a driving simulator, an RSU simulation device, an unmanned driving simulation vehicle, an intelligent driving database, and a data processing unit.

[0043] The driving simulator includes a parameter setting unit, a scene calculation unit, a real-time scene display unit, a GPS simulator, and a vehicle status transmission unit; the RSU simulation device includes a host and a radio frequency unit; the unmanned driving simulation car includes an on-board unit OBU, a vehicle data storage device, a positioning sensor, a decision-making planning module to be tested, a path parameter setting unit, and an Ethernet interface; the intelligent driving database includes a high-precision map library, a traffic rules database, and a scene database; the data processing unit includes a data storage device and a data analysis system.

[0044] The driving simulator is used to configure the initial state of the car and the static traffic environment of the driving area according to the parameters set by the user, calculate the real-time driving state of the vehicle and the dynamic traffic environment, and display the simulated driving scene in real time. The parameter setting unit is used for the user to set the simulated car driving area, the simulated car initial position, the simulated car model and size. The scene calculation unit generates the static traffic environment of the driving area according to the initial parameters set by the parameter setting unit and the intelligent driving database, initializes the state of the traffic lights at the intersection, initializes the dynamic electronic sign signs, and initializes the traffic state of the road section. During the simulation test, based on the traffic scene at the last moment, the position, speed, acceleration, and direction of the simulated car are dynamically calculated and updated in combination with the current state of the car, the static traffic environment, and the behavior decision results. A new simulated traffic scene is generated. The updated information is the current position, speed, acceleration, direction, and current simulated traffic scene of the simulated vehicle. The scene calculation unit sends the current position of the vehicle to the GPS simulator and sends the current traffic scene data to the real-time scene display unit. The road, environment, and roadside equipment data in the current traffic scene are sent to the RSU simulation device host, the updated position of the vehicle is sent to the GPS simulator, and the updated speed and direction information of the vehicle is sent to the vehicle state transmission unit. The vehicle state transmission unit transmits the vehicle size, speed, and direction information to the vehicle data storage device through the Ethernet interface of the unmanned driving simulation vehicle in an Ethernet wired manner, and the vehicle data storage device updates the vehicle real-time size, speed, and direction information therein. The real-time scene display unit displays the virtual scene of the road ahead on the display screen from the observation angle of the simulation vehicle based on the real-time scene data generated by the scene calculation unit, and updates it in real time.

[0045] The intelligent driving database provides the required high-precision map data, traffic scene data, and traffic rules data for the driving simulator and the unmanned driving simulation car decision-making and planning module. The scenario simulation database includes vehicle-road collaborative scenarios, which are used to simulate green wave speed guidance, front congestion reminders, road hazard reminders, speed limit sign reminders, construction sign reminders, red light running warnings, blind spot pedestrian crossing warnings, ramp vehicle merging warnings and other scenarios. The high-precision map library is static high-precision map data with lane-level accuracy, and the map covers an area of ​​no less than 100 square kilometers. The traffic rules database contains traffic regulations that motor vehicles, non-motor vehicles, and pedestrians should abide by when traveling on the road as stipulated by the national traffic management department.

[0046] The RSU simulation device is used to communicate with the OBU of the unmanned driving simulation car. The host generates real-time data of the roadside unit in the simulation scene based on the roadside equipment data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit. The radio frequency unit broadcasts RSU information in a wireless transmission manner and communicates with the on-board terminal OBU of the unmanned driving simulation car.

[0047] The unmanned driving simulation car is a simulation system with functions such as sensing traffic environment information, vehicle status, path planning and behavior decision-making. The OBU is an on-board unit device, which receives the road traffic environment information sent by the RSU simulator in a wireless transmission mode and saves this information in the vehicle data storage. The positioning sensor receives the current position information of the vehicle sent by the GPS simulator and updates the vehicle position information in the vehicle data storage at the same time. The vehicle data storage stores and updates the road traffic environment information, the current position, speed, driving direction and other information of the vehicle transmitted by the OBU in real time. The decision-making planning module to be tested is a decision-making planning module for testing performance. According to the current environment information and vehicle status, path planning and behavior decision-making are carried out, and the drive control, steering control, braking control and other signals required for the next operation of the vehicle are output, and these information are transmitted to the scene calculation unit of the driving simulator in a wired manner through the Ethernet interface. The path parameter setting unit is an interface for receiving user input parameters before the unmanned driving car carries out the decision-making planning test. The input parameters include parameters such as the end point, the must-pass point on the way, and priority conditions such as time, energy consumption, and distance.

[0048] The data storage device stores user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent by the GPS simulator to the positioning sensor, the length, width, height, position, speed, acceleration, direction and other data of the unmanned driving simulation car calculated by the vehicle state transmission unit and sent to the vehicle data storage device, the RSU information broadcast by the RSU simulation device, and the drive control, steering control and brake control signals sent by the decision-making and planning module to be tested to the scene calculation unit. The data analysis system is used to analyze the above stored data and evaluate the performance of the decision-making and planning module to be tested.

[0049] like Figure 2 As shown, a simulation test method for a decision-making planning module of an intelligent networked vehicle comprises the following steps:

[0050] 1) The user selects the driving area, initial position, model and size of the simulated car through the parameter setting unit of the driving simulator, selects weather parameters such as sunny, rainy or haze, and sets temporary construction sections and prohibited traffic conditions in the driving area as needed. The user selects the car's destination through the unmanned driving simulation car path parameter setting unit, enters the waypoints in sequence, and selects one condition from the priority conditions such as shortest time, least energy consumption, shortest distance, and least traffic lights as the priority basis for path planning.

[0051] 2) The scene calculation unit reads the model data in the scene simulation database, high-precision map library, and traffic rule library in the intelligent driving database according to the parameters set by the driving simulator parameter setting unit, and configures the parameters of the road, traffic environment, pedestrians, other vehicles, etc. in the driving area through calculation, initializes the parameters of the traffic facilities contained in the area, and generates a virtual scene. At the same time, the corresponding roadside unit RSU status data is configured, and the length, width, height, and initial position of the simulated car are configured. The current speed and acceleration of the car are 0, and the initial value of the direction is 0.

[0052] 3) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's length, width, height, speed, acceleration, and direction data to the vehicle state transmission unit.

[0053] 4) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the length, width, height, speed, acceleration, and direction data of the simulated car to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores it in the vehicle data storage. The GPS simulator sends the location data of the simulated car to the positioning sensor via wireless transmission and stores it in the vehicle data storage.

[0054] 5) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit. The RSU simulation device and the OBU of the simulated car exchange information. The radio frequency unit broadcasts the RSU status data by wireless transmission. After receiving it, the OBU stores the roadside unit status data in the vehicle data storage.

[0055] 6) After the user confirms the parameter settings, the decision-making planning module to be tested carries out global path planning. The decision-making planning module to be tested reads the data such as the car driving area, road environment, traffic facilities, initial position and size of the car in the vehicle data storage according to the user input parameters, and then combines the high-precision map library and traffic rules database in the intelligent driving database to carry out global path planning, and obtains a global path that meets the optimal conditions from the starting point, through the waypoints in sequence, and finally to the end point.

[0056] 7) If the current simulated car is in the initial stationary state, the decision-making planning module to be tested makes a decision to start the car. At this time, the speed and acceleration of the simulated car are both zero, and the position is the starting point. Otherwise, the decision-making planning module to be tested will make behavioral decisions based on the road environment, traffic facilities, traffic participant parameters, and car position and size data in the current scene stored in the vehicle data storage, combined with the high-precision map library and traffic rules database in the intelligent driving database, to determine the driving behavior that the simulated car should take next, and generate decision instructions for starting, lane keeping, lane changing, steering, obstacle avoidance or braking and other behaviors and duration.

[0057] 8) Based on the global path, according to the current speed, acceleration, position, direction, and size of the car, combined with the current traffic facilities and road environment parameters, the decision-making planning module to be tested carries out local path planning according to the decision-making instructions, calculates the driving trajectory and trajectory points of the unmanned car in the short future, determines the driving speed, acceleration, and direction of each trajectory point, and calculates and generates the driving control, steering control, braking control and other signals of the car. Afterwards, the decision-making planning module to be tested sends the driving control, steering control, braking control and other signal instructions, the vehicle's driving speed, acceleration, direction, position and other information at the next moment to the scene calculation unit through the Ethernet interface and Ethernet wired transmission.

[0058] 9) After receiving the instruction sent by the decision-making and planning module of the vehicle to be tested, the scene calculation unit updates the speed, acceleration, direction and position of the vehicle. Based on the traffic scene at the last moment, according to the model data in the scene simulation database, high-precision map library and traffic rule library in the intelligent driving database, and updates the current parameters such as roads, traffic environment, pedestrians, other vehicles, etc., updates the virtual scene, and configures the corresponding roadside unit RSU status data.

[0059] 10) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's speed, acceleration, and direction data to the vehicle state transmission unit.

[0060] 11) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the simulated car speed, acceleration, and direction data to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage. The GPS simulator sends the simulated car's location data to the positioning sensor via wireless transmission and stores them in the vehicle data storage. The vehicle data storage updates the simulated car's current speed, acceleration, direction, and location data.

[0061] 12) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit. The RSU simulation device sends the RSU status data to the OBU through the radio frequency unit, and the vehicle data storage device updates the roadside unit status data therein.

[0062] 13) The scene calculation unit determines whether the current simulated car collides with the static or dynamic obstacle. If a collision occurs, step 15 is executed. Otherwise, the next step is executed.

[0063] 14) The scene calculation unit determines whether the current simulated car has reached the end point. If it has reached the end point, execute step 16). Otherwise, return to step 7).

[0064] 15) When simulating a car collision, the behavior decision or local path planning performance of the decision-making planning module to be tested was defective and failed the test.

[0065] 16) The data storage of the data processing unit stores the user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent by the GPS simulator to the positioning sensor, the length, width, height, position, speed, acceleration, direction and other data of the simulated car sent by the vehicle state transmission unit to the vehicle data storage, the RSU real-time status information broadcast by the RSU simulation device, and the drive control, steering control and brake control signals sent by the decision-making and planning module to be tested to the scene calculation unit. The data analysis system analyzes the above stored data to evaluate the performance of the decision-making and planning module to be tested.

[0066] Embodiment 1

[0067] Taking the road construction warning scenario as an example, the implementation process of the intelligent connected vehicle decision-making planning module simulation system and simulation testing method of the present invention is explained.

[0068] Road construction warning scenarios such as Figure 3 As shown in the figure, the road is a long straight road with at least one lane, and a construction sign is set on the road section. The roadside unit RSU is installed on the construction sign, and the intelligent connected car drives towards the sign.

[0069] The intelligent connected car enters the right lane in advance, and the speed is at least 30km / h before 100m away from the construction area sign. When the distance between the intelligent connected car and the construction area sign is less than 100m, the on-board unit OBU will receive the construction area prompt information broadcast by the roadside unit (RSU) and, after processing, prompt the driver to enter the construction section ahead. The intelligent connected car slows down and bypasses the construction section before continuing forward. For this scenario, the requirements for the intelligent connected car decision-making and planning module to pass the performance test include:

[0070] (a) The intelligent connected vehicle slows down and does not come into contact with the construction obstacles.

[0071] (b) When the construction area becomes narrower, the intelligent connected vehicle replans its driving route and changes lanes to avoid the construction area.

[0072] (c) After passing the construction area, the intelligent connected vehicle returns to the original lane and continues to move forward.

[0073] In view of this scenario, the test method of the present invention is applied to carry out simulation testing of the decision-making planning module of intelligent connected vehicles, and the steps are as follows:

[0074] Step 1,

[0075] The intelligent connected vehicle decision-making and planning module simulation test system updates the traffic scenario to a road construction warning scenario. The road in this scenario is a long straight road with at least one lane, and a construction sign is set up on this section of the road. The roadside unit RSU is installed on the construction sign. The simulated car is driving forward at a speed of not less than 30km / h in the right lane. When the distance between the intelligent connected vehicle and the construction area sign is greater than 100m, the roadside unit RSU status data is empty. When the distance between the intelligent connected vehicle and the construction area sign is not greater than 100m, the roadside unit RSU status data contains information about the location area of ​​the construction section and the type of construction. The scene calculation unit configures the corresponding roadside unit RSU status data.

[0076] Step 2,

[0077] The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's speed, acceleration, and direction data to the vehicle state transmission unit.

[0078] Step 3,

[0079] The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the simulated car speed, acceleration, and direction data to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage. The GPS simulator sends the simulated car's location data to the positioning sensor via wireless transmission and stores them in the vehicle data storage. The vehicle data storage updates the simulated car's current speed, acceleration, direction, and location data.

[0080] Step 4,

[0081] The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit. The RSU simulation device sends the RSU status data to the OBU through the radio frequency unit, and the vehicle data storage updates the roadside unit status data therein.

[0082] Step 5,

[0083] The scene calculation unit determines whether the current simulated car collides with static or dynamic obstacles. If a collision occurs, the test ends and data analysis is then performed. Otherwise, the next step is executed.

[0084] Step 6,

[0085] The decision-making planning module to be tested makes behavioral decisions based on the road environment, traffic facilities, traffic participant parameters, and car location and size data in the current scene stored in the vehicle data storage, combined with the high-precision map library and traffic rules database in the intelligent driving database. Logical reasoning is performed on the current vehicle status and surrounding environment status to predict the risk of collision between the vehicle and the facilities in the construction area. The decision-making planning module to be tested forms decision instructions for starting, lane keeping, lane changing, steering, obstacle avoidance or braking and other behaviors and durations.

[0086] Step 7,

[0087] On the basis of the global path, according to the current speed, acceleration, position, direction, size of the car, combined with the current traffic facilities and road environment parameters, and according to the decision instructions, the decision planning module to be tested carries out local path planning. This generates the driving trajectory and trajectory points, determines the driving speed, acceleration and direction of each trajectory point, and calculates and generates the driving control, steering control, braking control and other signals of the car. After that, the decision planning module to be tested sends the driving control, steering control, braking control and other signal instructions, the vehicle's driving speed, acceleration, direction, position and other information at the next moment to the scene calculation unit through the Ethernet interface and Ethernet wired transmission.

[0088] Step 8,

[0089] After receiving the instruction sent by the decision-making planning module of the vehicle to be tested, the scene calculation unit updates the speed, acceleration, direction and position of the vehicle. Based on the traffic scene at the previous moment, the current road, traffic environment and other parameters are updated. The scene calculation unit updates the virtual scene and configures the corresponding roadside unit RSU status data. If the current simulated vehicle is less than 50 meters away from the boundary of the construction area, return to step 2. Otherwise, the road construction warning scene ends.

[0090] For the road construction warning scenario, the data analysis system of the simulation test system of the intelligent connected vehicle decision-making and planning module analyzes the high-precision positioning signal sent to the positioning sensor by the GPS simulator stored in the data storage device, the length, width, height, position, speed, acceleration, direction and other data of the unmanned driving simulation vehicle calculated by the vehicle state transmission unit and sent to the vehicle data storage device, the RSU information broadcast by the RSU simulation device, and the drive control, steering control and braking control signals sent by the decision-making and planning module to be tested to the scene calculation unit. When it is concluded that the simulated vehicle has no collision with the surrounding static and dynamic obstacles, complies with traffic rules and traffic regulations indicated by traffic signs and signs, and meets the following conditions in the lane keeping, lane changing and other stages, the decision-making and planning module to be tested is evaluated as passed.

[0091] In the lane keeping phase, the behavior decision and local path planning of the decision planning module to be tested must meet the following requirements:

[0092] a) The vehicle should travel in a straight line at a speed of not less than 20 km / h and at a steady speed;

[0093] b) The planned local path is along the lane line, and the maximum offset between the local path and the center line of the lane is no more than 0.3m;

[0094] c) The vehicle drives within its lane and the wheels do not touch the edge of the road.

[0095] During the lane change phase, the behavior decision and local path planning of the decision-making planning module to be tested must meet the following requirements:

[0096] a) Before changing lanes, the vehicle should travel in a straight line at a speed of no less than 20 km / h;

[0097] b) The driving route should not change two or more lanes continuously;

[0098] c) The lateral movement in the driving path should be completed continuously, and the path curvature should not have sudden changes;

[0099] d) The lateral acceleration of the vehicle during lane change should not exceed 4m / s 2 .

[0100] Embodiment 2

[0101] Furthermore, taking the blind spot pedestrian crossing warning scenario as an example, the implementation process of the intelligent connected vehicle decision-making planning module simulation system and simulation test method of the present invention is explained.

[0102] Blind spot pedestrian crossing warning scene Figure 4As shown. The road is a long straight road with at least two lanes, and a crosswalk is set in the road section. The roadside equipment RSU monitors the situation of pedestrians crossing the lane on the crosswalk through the camera and broadcasts the crossing information to surrounding vehicles. There is an obstacle between the car and the pedestrian on the crosswalk, which blocks the line of sight and creates a blind spot. The smart connected car drives towards the crosswalk, and the pedestrian crosses the road along the crosswalk at the same time, and there is a risk of collision between the two. The smart connected car has a speed of at least 30km / h before the crosswalk is 100m away. When the smart connected car is less than 50m away from the crosswalk, a pedestrian starts to cross the lane at a speed of 3km / h from about 20cm on the right side, and moves to the center of the lane and stops. The on-board unit OBU of the smart connected car receives the pedestrian information in the blind spot sent by the roadside equipment RSU, determines that a pedestrian has entered the blind spot, and immediately issues a blind spot warning to the driver. The smart connected car slows down until it stops in front of the pedestrian. The pedestrian continues to cross the road about 20cm to the left, and then the intelligent connected car passes the intersection at the original speed. For this scenario, the requirements for the intelligent connected car decision-making and planning module to pass the performance test include:

[0103] (a) After receiving the pedestrian crossing information sent by the roadside equipment, the intelligent connected vehicle slows down and stops.

[0104] (b) The intelligent connected vehicle stops before the stop line and the distance to the stop line is at least 1.5 meters.

[0105] (c) The intelligent connected vehicle does not collide with pedestrians.

[0106] (d) After the smart connected car and other pedestrians stop on the left side of the lane, they start and cross the crosswalk at the original speed.

[0107] In view of this scenario, the test method of the present invention is applied to carry out simulation testing of the decision-making planning module of intelligent connected vehicles, and the steps are as follows:

[0108] Step 1,

[0109] The intelligent connected vehicle decision-making and planning module simulation test system updates the traffic scenario to a pedestrian crossing warning scenario. The road in this scenario includes a long straight road with two lanes. Roadside units RSU are installed on the road ahead. Trees, street lights, buildings, etc. are arranged on both sides of the road. A crosswalk is set in the road ahead. There are many trees in the isolation belt between the two lanes, simulating that the car's line of sight in the direction of the crosswalk is blocked, resulting in a blind spot. The simulated car is driving forward at a speed of no less than 30km / h, and a pedestrian is about 20cm to the right of the lane and is stationary. The scene calculation unit is configured with the corresponding roadside unit RSU status data.

[0110] Step 2,

[0111] The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's speed, acceleration, and direction data to the vehicle state transmission unit.

[0112] Step 3,

[0113] The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car. The vehicle state transmission unit transmits the simulated car speed, acceleration, and direction data to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage. The GPS simulator sends the simulated car's location data to the positioning sensor via wireless transmission and stores them in the vehicle data storage. The vehicle data storage updates the simulated car's current speed, acceleration, direction, and location data.

[0114] Step 4,

[0115] The RSU simulation device host generates real-time data of the roadside unit in the simulation scene based on the roadside device data transmitted by the scene calculation unit. The RSU simulation device sends the RSU status data to the OBU through the radio frequency unit, and the vehicle data storage updates the roadside unit status data therein.

[0116] Step 5,

[0117] The scene calculation unit determines whether the current simulated car collides with static or dynamic obstacles. If a collision occurs, the test ends and data analysis is then performed. Otherwise, the next step is executed.

[0118] Step 6,

[0119] The decision-making planning module to be tested makes behavioral decisions based on the road environment, traffic facilities, traffic participant parameters, and car location and size data in the current scene stored in the vehicle data storage, combined with the high-precision map library and traffic rules database in the intelligent driving database. Logical reasoning is performed on the current vehicle status and surrounding environment status to predict the collision risk between the vehicle and pedestrians. The decision-making planning module to be tested forms decision instructions for starting, lane keeping, lane changing, steering, obstacle avoidance or braking and other behaviors and durations.

[0120] Step 7,

[0121] On the basis of the global path, according to the current speed, acceleration, position, direction, size of the car, combined with the current traffic facilities and road environment parameters, and according to the decision instructions, the decision planning module to be tested carries out local path planning. This generates the driving trajectory and trajectory points, determines the driving speed, acceleration and direction of each trajectory point, and calculates and generates the driving control, steering control, braking control and other signals of the car. After that, the decision planning module to be tested sends the driving control, steering control, braking control and other signal instructions, the vehicle's driving speed, acceleration, direction, position and other information at the next moment to the scene calculation unit through the Ethernet interface and Ethernet wired transmission.

[0122] Step 8,

[0123] After receiving the instruction sent by the decision-making planning module of the vehicle to be tested, the scene calculation unit updates the speed, acceleration, direction and position of the vehicle. Based on the traffic scene at the previous moment, when one of the following situations exists, the current road, traffic environment, pedestrian and other parameters are updated. The scene calculation unit updates the virtual scene and configures the corresponding roadside unit RSU status data. Return to step 2. Otherwise, execute step 9.

[0124] (1) If the simulated car is more than 50 m away from the crosswalk, the pedestrian remains stationary 20 cm to the right of the lane.

[0125] (2) If the simulated car is less than 50 m and more than 1.5 m away from the crosswalk, the pedestrian crosses the lane at a speed of 3 km / h from 20 cm to the right until it stops in the center of the lane.

[0126] (3) If the simulated car stops in front of the stop line and the pedestrian has not yet reached the center of the lane, the pedestrian continues to cross at a speed of 3 km / h until he moves to the center of the lane and stops.

[0127] (4) If the simulated car stops in front of the stop line and the pedestrian is stationary in the center of the lane, the pedestrian crosses at a speed of 3 km / h until it reaches 20 cm to the left of the lane and stops. During this period, the car maintains a speed of 0, an acceleration of 0, and a direction along the lane line, and its position remains unchanged.

[0128] (5) The simulated car stops in front of the stop line and the pedestrian stops 20 cm to the left of the lane.

[0129] Step 9: When the decision-making and planning module to be tested forms a start command, the simulated car starts to start from the front of the stop line, and the pedestrian crossing warning scene ends.

[0130] For the pedestrian crossing warning scenario, the data analysis system of the simulation test system of the intelligent connected vehicle decision-making planning module analyzes the high-precision positioning signal sent to the positioning sensor by the GPS simulator stored in the data storage device, the length, width, height, position, speed, acceleration, direction and other data of the unmanned driving simulation vehicle calculated by the vehicle state transmission unit and sent to the vehicle data storage device, the RSU information broadcast by the RSU simulation device, and the drive control, steering control and braking control signals sent by the decision-making planning module to be tested to the scene calculation unit. When it is concluded that the simulated vehicle has no collision with the surrounding static and dynamic obstacles, complies with traffic rules and traffic regulations indicated by traffic signs and signs, and meets the following conditions in the lane keeping stage and the braking stage, the decision-making planning module to be tested is evaluated as passed.

[0131] In the lane keeping phase, the behavior decision and local path planning of the decision planning module to be tested must meet the following requirements:

[0132] a) The vehicle should travel in a straight line at a speed of not less than 20 km / h and at a steady speed;

[0133] b) The planned local path is along the lane line, and the maximum offset between the local path and the center line of the lane is no more than 0.3m;

[0134] c) The vehicle drives within its lane and the wheels do not touch the edge of the road.

[0135] During the braking phase, the behavior decision and local path planning of the decision-making planning module to be tested must meet the following requirements:

[0136] a) The vehicle starts braking and decelerating when approaching the target parking point.

[0137] b) The vehicle stops before reaching the target parking point, and the distance from the parking position to the target parking point is no more than 1.5m.

Claims

1. A simulation test system for decision-making and planning modules of intelligent networked vehicles, characterized in that include: Driving simulator, RSU simulation equipment, unmanned driving simulation car, intelligent driving database, data processing unit; The driving simulator includes a parameter setting unit, a scene calculation unit, a real-time scene display unit, a GPS simulator, and a vehicle status transmission unit; the RSU simulation device includes a host and a radio frequency unit; the unmanned driving simulation car includes an on-board unit OBU, a vehicle data storage device, a positioning sensor, a decision-making planning module to be tested, a path parameter setting unit, and an Ethernet interface; the intelligent driving database includes a high-precision map library, a traffic rules database, and a scene database; the data processing unit includes a data storage device and a data analysis system.

2. According to claim 1, a simulation test system for intelligent connected vehicle decision-making and planning modules is characterized in that The driving simulator is used to configure the initial state of the car and the static traffic environment of the driving area according to the parameters set by the user, calculate the real-time driving state of the vehicle and the dynamic traffic environment, and display the simulated driving scene in real time; the parameter setting unit is used for the user to set the simulated car driving area, the simulated car initial position, the simulated car model and size; the scene calculation unit generates the static traffic environment of the driving area according to the initial parameters set by the parameter setting unit and the intelligent driving database, initializes the state of the traffic lights at the intersection, initializes the dynamic electronic signs and signs, and initializes the traffic state of the road section; during the simulation test, in combination with the current state of the car, the static traffic environment, and the behavior decision results, on the basis of the traffic scene at the previous moment, the position, speed, acceleration, and direction of the simulated car are dynamically calculated and updated, and a new simulated traffic scene is generated; the updated information is the current position, speed, acceleration, direction, and current simulated traffic scene of the simulated vehicle; the scene calculation unit sends the current position of the vehicle to the GPS simulator and sends the current traffic scene data to the real-time scene display unit; The road, environment, and roadside equipment data in the current traffic scene are sent to the RSU simulation device host, the updated vehicle position is sent to the GPS simulator, and the updated vehicle speed and direction information is sent to the vehicle state transmission unit. The vehicle state transmission unit sends the vehicle size, speed, and direction information to the vehicle data storage device via the Ethernet interface of the unmanned driving simulation car in an Ethernet wired manner. The vehicle data storage device updates the real-time size, speed, and direction information of the vehicle therein; the real-time scene display unit displays the virtual scene of the road ahead on the display screen from the observation angle of the simulated car based on the real-time scene data generated by the scene calculation unit, and updates it in real time.

3. The intelligent connected vehicle decision-making and planning module simulation test system according to claim 2 is characterized in that The intelligent driving database provides the required high-precision map data, traffic scene data, and traffic rules data for the driving simulator and the unmanned driving simulation car decision-making and planning module; the scene simulation database includes vehicle-road collaborative scenes, which are used to simulate green wave speed guidance, forward congestion reminders, road hazard reminders, speed limit sign reminders, construction sign reminders, red light running warnings, blind spot pedestrian crossing warnings, and ramp vehicle merging warning scenes; the high-precision map library is static high-precision map data with lane-level accuracy, and the map coverage area is not less than 100 square kilometers; the traffic rules database contains traffic regulations that motor vehicles, non-motor vehicles, and pedestrians should abide by when traveling on the road as stipulated by the national traffic management department.

4. The intelligent connected vehicle decision-making and planning module simulation test system according to claim 3 is characterized in that The RSU simulation device is used to communicate with the OBU of the unmanned driving simulation car. The host generates real-time data of the roadside unit in the simulation scene based on the roadside equipment data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit. The radio frequency unit broadcasts RSU information in a wireless transmission manner and communicates with the on-board terminal OBU of the unmanned driving simulation car.

5. The intelligent connected vehicle decision-making and planning module simulation test system according to claim 4 is characterized in that The unmanned driving simulation car is a simulation system with the functions of sensing traffic environment information, the vehicle's own state, and carrying out path planning and behavior decision-making; the OBU is an on-board unit device, which receives the road traffic environment information sent by the RSU simulator in a wireless transmission mode and saves the information into the vehicle data storage; the positioning sensor receives the current position information of the vehicle sent by the GPS simulator, and simultaneously updates the vehicle position information in the vehicle data storage; the vehicle data storage stores and updates in real time the road traffic environment information, the current position, speed, and driving direction information of the vehicle transmitted by the OBU; The decision-making planning module to be tested is a decision-making planning module used to test performance. It carries out path planning and behavior decision-making based on current environmental information and vehicle status, and outputs the driving control, steering control, and braking control signals required for the next operation of the vehicle. These information are transmitted to the scene calculation unit of the driving simulator in a wired manner through the Ethernet interface; the path parameter setting unit is an interface for receiving user input parameters before the unmanned vehicle carries out the decision-making planning test; the input parameters include the end point, parameters of the necessary points along the way, time, energy consumption, and distance priority conditions.

6. The intelligent connected vehicle decision-making and planning module simulation test system according to claim 5 is characterized in that The data storage device stores user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent to the positioning sensor by the GPS simulator, the length, width, height, position, speed, acceleration, and direction data of the unmanned driving simulation car calculated by the vehicle state transmission unit and sent to the vehicle data storage device, the RSU information broadcast by the RSU simulation device, and the drive control, steering control, and braking control signals sent to the scene calculation unit by the decision-making and planning module to be tested; the data analysis system is used to analyze the above stored data and evaluate the performance of the decision-making and planning module to be tested.

7. A simulation test method for decision-making and planning modules of intelligent networked vehicles, characterized in that The following steps are involved: 1) The user selects the driving area, initial position, model and size of the simulated car, selects weather parameters, and sets temporary construction sections and prohibited traffic conditions in the driving area through the parameter setting unit of the driving simulator; the user selects the car destination through the unmanned driving simulation car path parameter setting unit, inputs the waypoints in sequence, and selects one condition from the priority conditions of shortest time, least energy consumption, shortest distance, and least traffic lights as the priority basis for path planning; 2) The scene calculation unit reads the model data in the scene simulation database, high-precision map library, and traffic rule library in the intelligent driving database according to the parameters set by the driving simulator parameter setting unit, and configures the road, traffic environment, pedestrians, and other vehicle parameters of the driving area through calculation, initializes the parameters of the traffic facilities contained in the area, and generates a virtual scene; at the same time, configures the corresponding roadside unit RSU status data, configures the length, width, height, and initial position of the simulated car, the current speed and acceleration of the car are 0, and the initial value of the direction is 0; 3) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the length, width, height, speed, acceleration, and direction data of the car to the vehicle state transmission unit; 4) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car; the vehicle status transmission unit transmits the length, width, height, speed, acceleration, and direction data of the simulated car to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage; the GPS simulator sends the location data of the simulated car to the positioning sensor via wireless transmission, and stores them in the vehicle data storage; 5) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene according to the roadside device data transmitted by the scene calculation unit, and configures the channel and frequency for the radio frequency unit; the RSU simulation device and the OBU of the simulated car exchange information, and the radio frequency unit broadcasts the RSU status data in a wireless transmission mode. After receiving the data, the OBU stores the roadside unit status data in the vehicle data storage; 6) After the user confirms the parameter settings, the decision-making planning module to be tested carries out global path planning; The decision-making planning module to be tested reads the vehicle driving area, road environment, traffic facilities, initial position and size data of the vehicle in the vehicle data storage according to the user input parameters, and then combines the high-precision map library and traffic rules database in the intelligent driving database to carry out global path planning, and obtains a global path that meets the optimal conditions from the starting point, through the waypoints in sequence, and finally to the end point; 7) If the current simulated car is in the initial stationary state, the decision-making planning module to be tested makes a decision to start the car; at this time, the speed and acceleration of the simulated car are both zero, and the position is the starting point; Otherwise, the decision-making planning module to be tested makes behavioral decisions based on the road environment, traffic facilities, traffic participant parameters, and vehicle position and size data in the current scene stored in the vehicle data memory, combined with the high-precision map library and traffic rules database in the intelligent driving database, determines the driving behavior that the simulated vehicle should take next, and generates decision instructions; 8) On the basis of the global path, according to the current speed, acceleration, position, direction, and size of the car, combined with the current traffic facilities and road environment parameters, the decision-making planning module to be tested carries out local path planning according to the decision-making instructions, calculates the driving trajectory and trajectory points of the unmanned car in the short future, determines the driving speed, acceleration, and direction of each trajectory point, and calculates and generates the driving control, steering control, and braking control signals of the car; after that, the decision-making planning module to be tested sends the driving control, steering control, and braking control signal instructions, and the driving speed, acceleration, direction, and position information of the vehicle at the next moment to the scene calculation unit through the Ethernet interface in an Ethernet wired transmission mode; 9) After receiving the instruction sent by the decision-making and planning module of the vehicle to be tested, the scene calculation unit updates the speed, acceleration, direction and position of the vehicle; based on the traffic scene at the previous moment, according to the model data in the scene simulation database, high-precision map library and traffic rule library in the intelligent driving database, and updates the current road, traffic environment, pedestrians and other vehicle parameters, updates the virtual scene, and configures the corresponding roadside unit RSU status data; 10) The scene calculation unit transmits the RSU status data to the RSU simulation device host through Ethernet, transmits the virtual scene to the real-time scene display unit, transmits the simulated car position data to the GPS simulator, and transmits the car's speed, acceleration, and direction data to the vehicle state transmission unit; 11) The real-time scene display unit displays the road ahead, pedestrians, and environmental scenes from the perspective of the simulated car; the vehicle state transmission unit transmits the simulated car speed, acceleration, and direction data to the Ethernet interface of the unmanned simulated car via Gigabit Ethernet, and stores them in the vehicle data storage; the GPS simulator sends the simulated car's location data to the positioning sensor via wireless transmission, and stores them in the vehicle data storage; The vehicle data memory updates the current speed, acceleration, direction, and position data of the simulated vehicle; 12) The RSU simulation device host generates real-time data of the roadside unit in the simulation scene according to the roadside device data transmitted by the scene calculation unit; the RSU simulation device sends the RSU status data to the OBU through the radio frequency unit, and the vehicle data storage device updates the roadside unit status data therein; 13) The scene calculation unit determines whether the current simulated car collides with the static or dynamic obstacle. If a collision occurs, execute step 15); otherwise, execute the next step; 14) The scene calculation unit determines whether the current simulated car has reached the end point. If it has reached the end point, step 16) is executed; Otherwise, return to step 7); 15) When simulating a car collision, the behavior decision or local path planning performance of the decision-making planning module to be tested was defective and failed the test; 16) The data storage device of the data processing unit stores the user input parameters, the global path generated by the decision-making and planning module to be tested, the high-precision positioning signal sent by the GPS simulator to the positioning sensor, the simulated car length, width, height, position, speed, acceleration, and direction data sent by the vehicle status transmission unit to the vehicle data storage device, the RSU real-time status information broadcast by the RSU simulation device, and the drive control, steering control, and braking control signals sent by the decision-making and planning module to be tested to the scene calculation unit; the data analysis system analyzes the above stored data to evaluate the performance of the decision-making and planning module to be tested.

Images