Establishment and application research of a six-degree-of-freedom autonomous driving simulation test platform
[Objective]Driving simulators offer a realistic and highly repeatable human-vehicle-road simulation environment,which is crucial for studying active safety control and gathering experimental data.To enhance vehicle-assisted driving safety and comfort,an intelligent driving simulation test platform with a six-degree-of-freedom(6-DOF)motion platform has been established to investigate the human-vehicle-environment system.[Methods]The simulation test platform consists of four main components:a 6-DOF motion platform,a by-wire chassis and vehicle body,a visual simulation and projection system,and a real-time simulation system.The 6-DOF motion platform features a high response time of 15 ms,a wide angle range of±25°,and a substantial stroke of 2,550,020 mm.The by-wire chassis,controlled via the TTL protocol,includes an independent hydraulic braking system to replicate the brake pedal feel of a conventional vehicle.The visual simulation system employs a 180° curved large screen,displayed by four high-performance projectors,with edge blending managed by the HY-MAX-F-HX-CC210D high-definition edge processor.The real-time simulation system features a lower-level NI real-time host,control cabinet,and high-performance upper-level computer host.With SCANeR scenario simulation software,the platform can create custom driving scenarios and conditions,simulating autonomous driving,sensors,vehicle dynamics,and driver-in-the-loop conditions.[Results]The proposed apparatus has been utilized to develop autonomous driving takeover alert strategies and study driver attention allocation.By constructing various traffic flow scenarios using SCANeR software,38 licensed participants'response times to three remainder strategies were analyzed.Results indicated that in the absence of monitoring requests,the time to return hands to the steering wheel and assume control was considerably longer than scenarios with a request strategy.The takeover request weakly affected the time for the eyes to focus on the road.Monitoring requests effectively enhanced takeover efficiency,whereas excessive requests decreased driver readiness.Moreover,different alert sounds impacted driver takeover performance.Furthermore,the platform collected data on comfort,safety,and parking efficiency from eight drivers under different autonomous parking speed strategies,providing insights for selecting vehicle speeds in automated parking assist systems.[Conclusions]The autonomous driving simulation platform seamlessly integrates real vehicle controls like the gas pedal,brake pedal,and steering wheel with real-time simulation technology to accurately mimic vehicle dynamics.Incorporating actual human driving behaviors,it enables the development and testing of vehicle performance and simulation test bed technologies,offering comprehensive subjective and objective evaluations.The platform substantially enriches research methods for autonomous driving simulation,making it easier for researchers to understand the experimental environment and design well-informed experiments.
simulation driving experimentdriving simulatorreal-time simulation systemautonomous drivingtesting and evaluation
万方数据
维普
