Local Trajectory Planning And Safety Assessment On Autonomous Vehicles

Autonomous driving has become a hot topic in recent years.Several universities,research institutions,automotive enterprises and technology companies strive to have their place in this field.Planning is an important component in the study of autonomous vehicles.And the local trajectory planning is a pivotal issue,which will greatly influence the safety and comfort of the vehicle.Security problem is a fundamental issue for autonomous vehicles.The autonomous vehicle accidents occurred in the last few years make the public suspicious of driverless technology.To ensure the safety of the vehicles in different scenarios,the safety assessment part has attracted the attention of experts and scholars.First,we investigate the trajectory planning in static environments.We propose a novel A~* algorithm with equal-step sampling based on vehicle kinematic model to improve the comfort of the path as much as possible.The simulation results illustrate the feasibility of this approach on lane-keep and obstacle avoidance in different scenarios.In addition,the real vehicle test results of autonomous vehicle demonstrate the capability of real-time implementation.Second,the trajectory planning in dynamic environments is considered.To verify the capability of fully automated driving,an ambitious maneuver is the lane change.The motion prediction of other traffic participants is fully involved in the planning.The reference trajectory can be obtained by using high precision map(HD map)and lane detection.Then model predictive control(MPC)is utilized to optimize the reference trajectory according to the current state of autonomous vehicle.Different prediction horizons and coordinate transformation are adopted to optimize the planning.By doing so,it is easy to take the constraint conditions into account and the result is more intuitive.Finally,we handle the safety assessment algorithm.The Monte Carlo simulation is used to predict the probabilistic occupancy of the object,and prevent potential collisions sequentially.The probabilistic occupancy of other traffic participants are computed offline and the obtained result is then adopted in real-time application.Therefore,the real-time computational burden is reduced.The crash probability is put forward to verify the feasibility of real-time trajectory in safety assessment module.Two typical scenarios are analyzed: lane change on the straight road and turning at the intersection.The simulation results illustrate the efficiency of our method.