Automatic driving simulation system based on semi-physical simulation

Abstract

The invention relates to the field of automatic driving simulation testing, in particular to an automatic driving simulation system based on semi-physical simulation, which is characterized by comprising an upper computer server, a scene simulation server, a real-time injection server, a sensor signal simulation system, a controller module and a man-machine interaction module, the upper computer server is respectively connected with the scene simulation server and the real-time injection server, the scene simulation server is connected with the controller module, and the real-time injection server is sequentially connected with the sensor signal simulation system and the controller module. Through integrated upper computer software and an automatic test process, the test efficiency of the automatic driving system is improved, and the development period is shortened; due to the design of the multi-channel sensor signal simulation system, the system can be compatible with various communication protocols and sensor types, and the flexibility and expansibility of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving simulation testing, and specifically to an autonomous driving simulation system based on hardware-in-the-loop simulation. Background Art

[0002] Vehicle hardware-in-the-loop simulation technology is a real-time simulation technology that combines physical hardware and simulation software. This technology realizes the comprehensive investigation and verification of vehicle system performance by connecting physical components, such as physical sensors, actuators, controllers, etc., into the loop of the simulation system.

[0003] A hardware-in-the-loop simulation system generally consists of a simulation computer, physical effect devices, tested physical objects, simulation devices, interface devices, and a support service system. It uses simulation devices to construct a physical environment for the object under test and conducts simulation tests together with physical models and mathematical models. In this system, physical objects are embedded in the software environment through computer interfaces, and both software and hardware need to run in real time to simulate the operating state of the entire vehicle system.

[0004] However, the current ordinary vehicle hardware-in-the-loop simulation system is too complex to operate, has poor user interactivity and flexibility, and cannot fully simulate complex situations in the real environment. Improvements are still needed for the above deficiencies. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an autonomous driving simulation system based on hardware-in-the-loop simulation, which can effectively solve the problems raised in the above background art.

[0006] To solve the above problems, the technical solution adopted by the present invention is: an autonomous driving simulation system based on hardware-in-the-loop simulation, which is characterized by including a host computer server, a scenario simulation server, a real-time injection server, a sensor signal simulation system, a controller module, and a human-computer interaction module. The host computer server is respectively connected to the scenario simulation server and the real-time injection server. The scenario simulation server is connected to the controller module. The real-time injection server is sequentially connected to the sensor signal simulation system and the controller module; The host computer server is used for engineering configuration, experiment management, and automated testing; The scenario simulation server is used for generating a virtual 3D scenario, and sending real-time images to the controller module through a camera injection system to simulate real camera input; The real-time injection server includes a dynamics model and interface conversion. The dynamics model is used for calculating vehicle motion, and the interface conversion is used for converting dynamics data into in-vehicle protocol signals; The sensor signal simulation system is used for receiving control instructions from the controller module and sending chassis signals and environmental signals to the controller module; The controller module is used to run autonomous driving algorithms, including positioning algorithms, perception algorithms, and planning and control algorithms; The human-machine interaction module includes an input device and a display screen. The input device includes a steering wheel, an accelerator, and a brake pedal.

[0007] Preferably, the engineering configuration in the host computer server includes setting vehicle parameters, sensor types, and communication protocols; the experiment management includes selecting test scenarios, controlling the start and stop of tests, and recording data; the automated testing includes batch execution of test cases and automatic report generation.

[0008] Preferably, the virtual 3D scene in the scenario simulation server includes roads, vehicles, and pedestrians.

[0009] Preferably, the control instructions in the sensor signal simulation system include brake signals, the chassis signals include vehicle speed and gear position, and the environmental signals include simulated radar and LiDAR data.

[0010] Preferably, the positioning algorithm in the controller module is used to combine virtual GPS and IMU data; the perception algorithm is used to process camera images and radar point clouds; the planning and control algorithm is used to generate steering, acceleration, or braking instructions.

[0011] Preferably, the display screen in the human-machine interaction module is used to display the position of the vehicle, the perception results, and the driving path in the virtual scene in real time.

[0012] Compared with the prior art, the present invention provides an autonomous driving simulation system based on hardware-in-the-loop simulation, which has the following beneficial effects: Improved test efficiency: Through the integrated host computer software and automated test process, the present invention significantly improves the test efficiency of the autonomous driving system and shortens the development cycle; Enhanced system compatibility: The design of the multi-channel sensor signal simulation system enables the present invention to be compatible with various communication protocols and sensor types, enhancing the flexibility and scalability of the system; Enhanced system safety: The integration of the real-time dynamics model and the interface model helps to discover and solve potential safety problems in the development stage in advance, improving the overall safety of the autonomous driving system; Optimized user experience: The user interaction and visualization observation functions enable users to observe and debug the system more conveniently, improving the user experience and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic structural diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0015] Referring to Figure 1 , the present invention provides an autonomous driving simulation system based on hardware-in-the-loop simulation, and its composition and functions are as follows: The system consists of the following modules working together: Host computer server There is host computer software connected to the host computer server; Engineering configuration: Set vehicle parameters (mass, dimensions), sensor types (camera / radar), communication protocols (CAN / CANFD, etc.); Experiment management: Select test scenarios (such as intersections), control the start and stop of the test, and record data; Automated testing: Batch execute test cases (such as emergency braking) and automatically generate reports.

[0016] 2. Scenario simulation server Generate 3D scenarios such as virtual roads, vehicles, and pedestrians, and send real-time images (such as 1080P video streams) to the controller module through the camera injection system to simulate real camera inputs.

[0017] 3. Real-time injection server Dynamics model: Calculate vehicle motion (such as speed, steering angle); Interface conversion: Convert dynamics data into in-vehicle protocol signals (such as CAN messages).

[0018] Sensor signal simulation system Receive control instructions (such as brake signals) from the controller module, and send chassis signals (vehicle speed, gear) and environmental signals (simulated radar / LiDAR data) to the controller module.

[0019] Controller module Run the autonomous driving algorithm: Positioning (combining virtual GPS and IMU data); Perception (processing camera images and radar point clouds); Planning and control (generating steering and acceleration / brake instructions).

[0020] Human-machine interaction module The steering wheel, accelerator / brake pedals are used as input devices, and the user's operations directly affect the vehicle behavior in the simulation; The display screen shows the position of the vehicle in the virtual scene, the perception results, and the driving path in real time.

[0021] As a specific embodiment of the present invention: The system operation process is as follows: Step 1, Initialization configuration: Set vehicle parameters, select test scenarios, and load communication protocols; Step 2, Closed-loop testing: The scenario simulation server generates images and radar signals → The controller module processes the signals and outputs control instructions → The sensor signal simulation system transmits the instructions to the dynamics model → The dynamics model updates the vehicle state → The new state is fed back to the controller module to form a closed loop; Step 3, Dynamic intervention: The user can control the vehicle in real time through the steering wheel / pedals, or modify parameters (such as road slipperiness) in the host computer server, and the system responds immediately; Step 4, Result display: The display screen synchronously shows the vehicle movement trajectory, obstacle recognition results (such as pedestrians marked with red boxes), and algorithm decision-making processes (such as lane-changing paths).

[0022] The core implementation conditions of the system are: hardware, software, and synchronization; Hardware: A real-time computing server (ensuring millisecond-level response) and a multi-protocol communication interface (supporting CAN / CANFD, etc.) are required; Software: A dynamics model (such as vehicle motion equations), a virtual scene engine (such as 3D rendering), and a protocol conversion tool; Synchronization: The clocks of each module need to be strictly synchronized to ensure consistent data timing (for example, image and radar data time alignment).

[0023] The innovation points of the present invention are as follows: 1. Integrated host computer software design The host computer software in the present invention integrates three major functional modules: engineering configuration, experiment management, and automated testing, forming a highly integrated control center. This innovative design not only simplifies the operation process but also improves the overall efficiency and flexibility of the system. Compared with the dispersed software modules in the prior art, it realizes seamless connection of functions and real-time sharing of data.

[0024] 2. Multi-channel sensor signal simulation system The sensor signal simulation system in the present invention can receive control signals from the controller module and send sensor signals and chassis signals to the controller module through multiple communication protocols (such as CAN, CANFD, LIN, FlexRay); this multi-channel design enhances the compatibility and expandability of the system, enabling the present invention to simulate more complex and real driving scenarios.

[0025] 3. Integration of real-time dynamics model and interface model The dynamic model and interface model running in the real-time injection server are another major innovation of the present invention. These models can simulate the dynamic behavior of the vehicle and interface interactions in real time, providing a more accurate simulation environment for the controller module, which helps to detect and solve potential problems in advance during the development stage and improve the safety and reliability of the autonomous driving system.

[0026] 4. User Interaction and Visual Observation The present invention displays a reasonable perspective through the screen, enabling users to observe information such as the vehicle state, environmental changes, and controller output during the simulation process in real time. This user interaction and visual observation function not only improves the transparency and traceability of the test but also facilitates users to conduct problem diagnosis and optimization adjustments.

[0027] Advantages of the Present Invention: I. User Interaction and Flexibility Significantly enhances user interaction. By receiving the user's input signals in real time (such as throttle, brake, and steering wheel operations), users can directly participate in the simulation process. This instant feedback mechanism greatly improves the realism of the simulation and the user experience. In addition, users can easily adjust vehicle parameters (such as vehicle speed, acceleration, steering sensitivity, etc.) and scene settings (such as weather conditions, road types, traffic conditions, etc.) through the upper computer component, achieving a highly customizable simulation environment. This flexibility not only meets diverse test requirements but also improves the pertinence and efficiency of the test.

[0028] In contrast, ordinary vehicle hardware-in-the-loop simulation systems have limitations in terms of user interaction and flexibility. Although they can also provide a certain degree of parameter adjustment and scene setting functions, they are often complex to operate and have a limited adjustment range. In addition, ordinary systems usually lack the ability to receive user input signals in real time, resulting in relatively weak user experience and simulation realism.

[0029] Testing and Verification Capabilities Has significant advantages in terms of testing and verification capabilities. The positioning algorithm, perception algorithm, planning and control algorithm running in the controller module can be fully verified in the simulation environment. The sensor signal simulation system can simulate the output signals of various sensors and chassis signals, providing comprehensive and accurate data support for the testing of the controller module. This comprehensive testing and verification capability ensures the accuracy and robustness of the algorithm in different scenarios, providing a strong guarantee for the research and development of autonomous driving technology.

[0030] Ordinary vehicle hardware-in-the-loop simulation systems can also provide certain support in terms of testing and verification capabilities, but they are often limited to specific scenarios and parameter ranges. In addition, due to the limitations of sensor signal simulation, ordinary systems may not be able to fully simulate complex situations in the real environment, thus affecting the accuracy and reliability of the test results.

[0031] Visualization and Monitoring A reasonable perspective is displayed through the screen, enabling users to observe in real time information such as the vehicle state, environmental changes, and controller outputs during the simulation process. This visualization function not only facilitates users to conduct problem diagnosis and optimization adjustments but also improves the transparency and traceability of the test. In addition, the experiment management software in the host computer software is responsible for recording key data during the simulation process, facilitating subsequent analysis and report generation.

[0032] Ordinary vehicle hardware-in-the-loop simulation systems have certain deficiencies in visualization and monitoring. Although they can also provide a certain degree of data recording and display functions, they often lack intuitiveness and ease of use. In addition, the capabilities of ordinary systems in data recording and analysis are also relatively limited, making it difficult to meet the requirements of complex test scenarios.

[0033] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An autonomous driving simulation system based on hardware-in-the-loop simulation, characterized in that It includes a host computer server, a scenario simulation server, a real-time injection server, a sensor signal simulation system, a controller module, and a human-machine interaction module. The host computer server is respectively connected to the scenario simulation server and the real-time injection server. The scenario simulation server is connected to the controller module. The real-time injection server is sequentially connected to the sensor signal simulation system and the controller module; The host computer server is used for engineering configuration, experiment management, and automated testing; The scenario simulation server is used to generate a virtual 3D scenario, send real-time images to the controller module through the camera injection system, and simulate real camera input; The real-time injection server includes a dynamics model and interface conversion. The dynamics model is used to calculate vehicle motion, and the interface conversion is used to convert dynamics data into in-vehicle protocol signals; The sensor signal simulation system is used to receive control instructions from the controller module and send chassis signals and environment signals to the controller module; The controller module is used to run autonomous driving algorithms, including positioning algorithms, perception algorithms, and planning and control algorithms; The human-machine interaction module includes an input device and a display screen. The input device includes a steering wheel, an accelerator, and a brake pedal.

2. The semi-physical simulation-based autonomous driving simulation system according to claim 1, wherein, The engineering configuration in the host computer server includes setting vehicle parameters, sensor types, and communication protocols; the experiment management includes selecting test scenarios, controlling test start and stop, and recording data; the automated testing includes batch execution of test cases and automatic report generation.

3. The self-driving simulation system based on hardware-in-the-loop simulation according to claim 1, characterized in that, The virtual 3D scenario in the scenario simulation server includes roads, vehicles, and pedestrians.

4. A semi-physical simulation-based autonomous driving simulation system according to claim 1, characterized in that, The control instructions in the sensor signal simulation system include brake signals. The chassis signals include vehicle speed and gear position. The environment signals include simulated radar and LiDAR data.

5. A semi-physical simulation-based autonomous driving simulation system according to claim 1, characterized in that, The positioning algorithm in the controller module is used to combine virtual GPS and IMU data; the perception algorithm is used to process camera images and radar point clouds; the planning and control algorithm is used to generate steering, acceleration, or braking instructions.

6. The semi-physical simulation-based autonomous driving simulation system according to claim 1, wherein The display screen in the human-machine interaction module is used to display the position of the vehicle, perception results, and driving path in the virtual scenario in real time.

Images