Development Of XIL Testing System Supporting Hardware,Driver And Vehicle And Algorithm Design

The Vehicle-to-Everything(V2X)technology is a comprehensive technology that integrates various transportation elements such as vehicles and roadside infrastructure,aiming to enhance the information acquisition capabilities of vehicles in driving scenarios.Prior to the production application of V2 X devices,rigorous testing is required.Traditional testing approaches based on physical vehicle testing are timeconsuming and economically expensive.The application of virtual simulation testing systems can simulate various application scenarios of V2 X devices,thereby improving the efficiency of algorithm development.It is currently a focal point of academic research.However,existing testing systems only focus on testing a specific in-vehicle environment,resulting in a limited approach that fails to meet the requirements of algorithm development tasks.To address this issue,this paper develops an XIL test system supporting hardware,driver and vehicle-in-the-loop and the corresponding application algorithms.The main contributions of this paper are as follows:Firstly,this paper proposes a XIL test system that supports hardware,driver and vehicle in the loop,and completes the construction and development of the test system hardware equipment.The test system adopts a modular architecture,including an upper computer,a driving simulator,a test stand and a test prototype,etc.The three in-loop test functions are realised through the collaboration of multiple modules,and the design and development of the underlying modules and the mapping of virtual test scenarios to real scenarios are completed.Compared with a single in-loop testing system,this system has more diverse testing methods,improves the efficiency of testing and meets the development needs of the algorithms for Telematics applications.Secondly,this paper investigates the application algorithms for V2 X,improving two typical application algorithms for forward collision warning and collaborative lane changing in response to realistic vehicle driving behaviour,road environments and complex traffic conditions.This paper presents a forward collision warning algorithm based on collision tolerance.By continuously comparing the estimated collision time of the vehicle at the current moment with the predetermined safety time,an instantaneous risk level is obtained.Subsequently,collision tolerance is calculated using an accumulation method,and the triggering timing of the warning is determined based on the collision tolerance.This approach reduces the impact of fluctuating speeds of the preceding vehicles on the accuracy of the algorithm,enhances the smoothness of the warning timing,and reduces the probability of false alarms.Furthermore,a cooperative lane change algorithm based on decoupling in the longitudinal and lateral directions is proposed.In the longitudinal direction,a cost function based on time,velocity,and acceleration is constructed,and the longitudinal velocity is obtained by solving the displacement-time curve function.This improves the real-time performance of the algorithm.In the lateral direction,a discretization method is employed to sample the planning space for vehicle lane change.The discrete trajectory is obtained through quadratic programming,enhancing the adaptability of the algorithm.Consequently,the algorithm can dynamically plan the desired lane change trajectory in real-time while satisfying the dynamic constraints.Finally,to validate the effectiveness of the proposed testing system and application algorithms,multiple experiments are conducted by integrating the designed application algorithms into the XIL testing system.The experimental results demonstrate that the proposed algorithms exhibit high accuracy and perform well under multiple sets of parameters in the testing scenarios,confirming the feasibility and effectiveness of the proposed algorithms.The paper also calculates the latency performance of the algorithms in practical applications and simulation testing conditions,compares the warning triggering processes in virtual and real environments,and verifies the effectiveness of the developed testing system.