The Decision Planning And Tracking Control Of Active Obstacle Avoidance For Intelligent Vehicle In Urban Road Scenarios

With the continuous increase in the number of automobiles in our country,the endless traffic accidents have attracted people’s attention.Active safety systems and advanced driver assistance systems in intelligent vehicles play a significant role in reducing the incidence of trafic accidents,and are considered to be the perfect way to solve traffic accidents.However,the existing research mainly focuses on highway scenes,and there are few studies on complex urban scenes.For this reason,this paper takes the active obstacle avoidance system of intelligent vehicles for urban scenes as the research object,and conducts research include decision,path planning,and tracking control.The main research contents are as follows:(1)The decision model of intelligent vehicle active obstacle avoidance is constructed.By analyzing the obstacle avoidance behavior of vehicles in urban traffic environment,which are divided into three basic behaviors:following,active braking and active steering.The safety distance model is established,and the main factors affecting vehicle obstacle avoidance are analyzed.The ego vehicle speed coefficient and the vehicle distance expectation coefficient are introduced.Under the fuzzy rules of expert experience,the implementation willingness of obstacle avoidance is the output,and an active obstacle avoidance decision model based on fuzzy logic reasoning is established.The feasibility of the decision model is verified on the cosimulation platform.(2)The path planning of vehicle active obstacle avoidance was studied.Firstly,convert the Cartesian coordinate system and the Frenet coordinate system,and sampling on the discrete roads.Secondly,the Hybrid A*algorithm is used to solve a convex optimization space domain that conforms to the vehicle kinematics,and the objective function is designed.The path is fitted by the fixed function method.Finally,the propose path planner was verified on the cosimulation platform based on the actual urban scene,and the results show that is safe and reliable.(3)A vehicle active obstacle avoidance path tracking controller is designed.On the basis of vehicle kinematics and dynamics,a vehicle tracking controller based on model prediction is designed,and it is verified and analyzed on a co-simulation platform with other common controllers under different vehicle speeds and road conditions.The results show that the designed lateral tracking controller based on model prediction has extremely high robustness to vehicle speed and road surface,and can maintain satisfactory tracking performance in various working conditions,and the control is stable,which is suitable for urban roads.Obstacle avoidance tracking control in the scene.(4)Build a simulation experiment platform.The effectiveness and reliability of obstacle avoidance decision,path planning and tracking control are verified by building an urban environmental on Prescan-Simulink-Carsim co-simulation platform.The simulation results show that the decision,planning,and control method described in this paper can perform safe and reliable active obstacle avoidance in urban environment scenarios(including going straight,no-signal T-junction,no-signal intersection).