Hardware-in-the-loop simulation test platform for automatic driving development

Abstract

The utility model relates to the technical field of automobile automatic driving simulation test, and specifically discloses a hardware-in-the-loop simulation test platform for automatic driving development. The platform comprises a simulation environment, a simulation module and automatic driving vehicle hardware, the simulation module is one or more vehicle dynamics simulation modules and is embedded in the simulation environment; the automatic driving vehicle hardware is single automatic driving function hardware or a hardware system composed of a plurality of automatic driving function hardware; the automatic driving vehicle hardware is connected with the simulation environment and the simulation module, the simulation environment outputs a simulated environment signal as sensing data input of the simulation module, and a processing result of the simulation module is transmitted to the automatic driving vehicle hardware, so that the automatic driving vehicle hardware generates a corresponding hardware signal and feeds the hardware signal into the simulation environment. According to the utility model, hardware of a real automatic driving vehicle, a simulation environment and a simulation module are integrated into a complete hardware-in-the-loop simulation test platform, and an automatic driving function can be effectively verified.

Description

Technical Field

[0001] The utility model relates to the technical field of automobile automatic driving simulation testing, in particular to a hardware-in-the-loop simulation testing platform. Background Art

[0002] With the deepening of the trend of automotive intelligence, the demand for testing automotive automation systems is growing rapidly, especially for the testing of L3 and above autonomous driving functions, which requires a huge amount of test mileage to ensure safety.

[0003] Simulation testing is widely adopted in the industry due to its advantages of being fast and low-cost, accounting for the absolute majority of the total test mileage (currently about 99% of the industry's autonomous driving test mileage is completed by simulation testing).

[0004] Existing simulation tests utilize hardware-in-the-loop (HIL) technology, where a real vehicle controller is connected to a simulation module to test the vehicle controller. However, existing HIL simulations only directly connect the simulation module to the vehicle controller hardware for simulation testing, lacking interactive data with the simulation environment, which in turn hinders the effective verification of autonomous driving functions. Utility Model Content

[0005] The utility model integrates the hardware of a real autonomous driving vehicle with the simulation environment and simulation modules into a complete hardware-in-the-loop simulation test platform, which forms a ring signal flow and can effectively verify the autonomous driving function.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a hardware-in-the-loop simulation test platform for autonomous driving development, which includes: a simulation environment, a simulation module and autonomous driving vehicle hardware;

[0007] The simulation module is one or more vehicle dynamics simulation modules embedded in the simulation environment;

[0008] Autonomous driving vehicle hardware refers to a single piece of autonomous driving functional hardware or a hardware system consisting of multiple pieces of autonomous driving functional hardware;

[0009] The autonomous driving vehicle hardware is connected to the simulation environment and the simulation module. The simulation environment outputs simulated environmental signals as perception data input of the simulation module. The processing results of the simulation module are further transmitted to the autonomous driving vehicle hardware as the hardware carrier of the simulation module, so that the autonomous driving vehicle hardware generates corresponding hardware signals and feeds them into the simulation environment.

[0010] Preferably, the single autonomous driving functional hardware includes: perception sensors, vehicle controllers, and vehicle power and transmission hardware.

[0011] Preferably, the perception sensors include: RGB camera, laser radar, infrared camera, millimeter wave radar and ultrasonic radar.

[0012] The utility model has the following beneficial effects:

[0013] This utility model integrates the hardware of a real autonomous driving vehicle with the simulation environment and simulation modules into a complete hardware-in-the-loop simulation test platform, forming a ring signal flow for testing, which can effectively verify the autonomous driving function of the hardware of the real autonomous driving vehicle being tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a structural diagram of an embodiment of the present utility model. DETAILED DESCRIPTION

[0015] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] like Figure 1 As shown, an embodiment of the present invention provides a hardware-in-the-loop simulation test platform for autonomous driving development, including: a simulation environment, a simulation module, and autonomous driving vehicle hardware.

[0017] The simulation environment is presented in the form of computer software and contains comprehensive information for guiding driving (generally referred to as high-precision maps in the autonomous driving industry), including but not limited to centimeter-level road structure information (such as road shape, slope, curvature, paving, direction, etc.), lane markings, road signs, traffic lights, and other important roadside object information (such as overhead objects, guardrails, trees, road edge types, roadside landmarks, etc.). The simulation environment can utilize prior art available before the filing date of this utility model, and this utility model is not limited to such information.

[0018] The simulation module is a vehicle dynamics simulation module (autonomous driving algorithm) corresponding to one or more mass-produced vehicle models. It is presented as computer software and embedded in the simulation environment. Embedding means that the simulated vehicle corresponding to the vehicle dynamics simulation module can move within a controlled, specific area of the simulation environment.

[0019] The simulation module may adopt the existing technology before the application date of the present invention. For example, the simulation module may adopt the commonly used commercial modules CarSim, CarMaker, Simulink, etc. in the industry. The present invention does not limit this.

[0020] Autonomous driving vehicle hardware refers to a single piece of autonomous driving functional hardware or a hardware system composed of multiple pieces of autonomous driving functional hardware. A single piece of autonomous driving functional hardware includes various perception sensors (such as RGB cameras, LiDAR, infrared cameras, millimeter-wave radar, ultrasonic radar, etc.), various vehicle controllers, and various vehicle power and transmission hardware.

[0021] The autonomous driving vehicle hardware is connected to the simulation environment and simulation module. The simulation environment outputs simulated environmental signals as perception data input for the simulation module. The processing results of the simulation module are further transmitted to the autonomous driving vehicle hardware, which serves as the hardware carrier of the simulation module, so that the autonomous driving vehicle hardware generates corresponding hardware signals and feeds them into the simulation environment (manifested as the simulated vehicle corresponding to the simulation module in the simulation environment is controlled by this signal to produce changes in vehicle dynamics behavior, such as moving forward, turning, etc.); the signal flow of this loop system can continue to operate for a specific length of time, thus forming a complete hardware-in-the-loop system for autonomous driving testing. This signal flow is manifested as the simulated vehicle corresponding to the vehicle dynamics module can continue to move for a specific time range in a controlled specific area of the simulation environment without being interrupted by non-human factors.

[0022] exist Figure 1 In the illustrated embodiment, the simulation environment is generated by 3D reconstruction of real road data. The real road data is road data collected by real sensors and integrated into the simulation environment using a data fusion algorithm. The data fusion algorithm can be based on prior art prior to the filing date of this utility model, and this utility model is not limited thereto. The 3D reconstruction can utilize existing technologies such as Structure from Motion (SfM) and Multi-View Stereo (MVS). The simulation environment has an editing function, enabling editing of elements of the simulation environment (such as traffic participants such as vehicles and pedestrians, weather, road markings, and traffic signs).

[0023] exist Figure 1 In the embodiment shown, the simulation environment is generated based on data collected from real roads, and the elements of the simulation environment can also be edited, so that the simulation test is close to the field test and the real road test in terms of realism and simulation consistency, and has the advantages of high authenticity, editability, strong flexibility and scalability, and easy reuse; Figure 1 The simulation test platform of the illustrated embodiment can overcome the defect that the existing simulation test platform is mainly based on the test environment (simulation environment and simulation module) built by software, and has obvious differences from field testing and real road testing in terms of realism and simulation consistency.

[0024] The above embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Any equivalent substitution or modification made by those skilled in the art based on the present invention shall fall within the protection scope of the present invention.

Claims

1. A hardware-in-the-loop simulation test platform for autonomous driving development, characterized by: It includes: Simulation environment, simulation modules, and autonomous vehicle hardware; The simulation module is one or more vehicle dynamics simulation modules embedded in the simulation environment; Autonomous driving vehicle hardware refers to a single piece of autonomous driving functional hardware or a hardware system consisting of multiple pieces of autonomous driving functional hardware; The autonomous driving vehicle hardware is connected to the simulation environment and the simulation module. The simulation environment outputs simulated environmental signals as perception data input of the simulation module. The processing results of the simulation module are further transmitted to the autonomous driving vehicle hardware as the hardware carrier of the simulation module, so that the autonomous driving vehicle hardware generates corresponding hardware signals and feeds them into the simulation environment.

2. A hardware-in-the-loop simulation test platform for autonomous driving development as claimed in claim 1, characterized in that: The single autonomous driving function hardware includes: perception sensors, vehicle controllers and vehicle power and transmission hardware.

3. A hardware-in-the-loop simulation test platform for autonomous driving development as claimed in claim 2, characterized in that: The perception sensors include: RGB camera, laser radar, infrared camera, millimeter wave radar and ultrasonic radar.

Images