Study On The Adaptive Cruise Control Methods Of Vehicle Under Different Driving Requirements

Adaptive cruise control(ACC)system can effectively reduce the workload of the driver by adjusting the velocity and distance of vehicle automatically according to the environment around the vehicle.It can improve the intelligence,operating efficiency and safety of the highway system.However,with the development of unmanned driving and intelligent networked vehicles,higher control accuracy is necessary for the ACC design to expand its application.So,it is necessary to design the ACC algorithm with these requirements such that it can promote the popularization of ACC system and the research of unmanned driving and connected vehicles.This paper studies the ACC algorithm designation problems with uncertain driving resistance,multi-spacing strategies,multi-objective requirements and different communication topology based on the nonlinear vehicle model.The main contributions are as follows:1.An ACC algorithm based on radial basis function neural network approximation is proposed to handle the uncertain driving resistance.First,the driving resistance of the vehicle is analyzed,and the dynamic model of the vehicle is reconstructed through the radial basis neural network;secondly,the ACC algorithm is designed based on the backstepping control.A virtual controller is designed to convert the distance control into velocity control.The neural parameters are estimated by designing an adaptive law.A nonlinear filter is designed to improve control accuracy and reduce the coupling relationship between the distance control and the velocity control.The designed algorithm uses the state information of the vehicle to learn the uncertain driving resistance such that the accuracy vehicle model is released.2.A model free robust ACC algorithm is designed for different space strategies.First of all,a virtual controller is designed to convert the vehicle spacing control into speed control by using the backstepping technology such that it can avoid the coupling of speed and spacing control brought by the speed-dependent spacing strategy.Then,a data-based coupling sliding mode surface and state observer were constructed to compensate for the complex nonlinear dynamics of vehicle,discrete errors and external disturbances such that the robustness of the control algorithm was improved.Finally,a hybrid ACC method including feedback control and robust control was designed.The designed algorithm only needs the input/output information of the vehicle(control input variation,distance and velocity),which can realize the accurate and robust conol for different spacing strategies.At the same time,it can also achieve the accurate and robust control under low sampling frequency.3.A prescribed peformanc velocity and distance switching ACC algorithm is proposed.If the follower is far away from its preceding vehicle,the prescribed peformanc velocity controller is used to adjust the velocity of the follower to the prescribed velocity.If the follower is close to its preceding vehicle,the prescribed peformanc distance controller is used to adjust the distance between the follower and its preceding vehicle.The prescribed error function is used to expand the switching interval and improve the robustness of the switching process.The designed controller is a model free output error feedback controller with low complexity.The control gain is adjusted to the actual required value through the time-varying adaptive law,and releases the necessary for the accurate vehicle model.It can ensure the velocity and spacing objectives in any case,and ensure that the spacing error changes within the expected range to avoid collision.4.A cooperative ACC algorithm is proposed using barrier function.First,the topology of the intelligent networked vehicle is described through the graph theory,and the distance error of the vehicle platoon is defined and constrained to the preset error range.Then,the barrier Lyapunov function is designed for the development of and the virtual controller and the real controller,where the disturbance terms such as vehicle nonlinear driving resistance and intelligent network topology switching disturbance are regarded as "like-disturbance" terms,which are compensated by designing an adaptive compensator.The designed controller does not need to establish an accurate model of nonlinear driving resistance,and has no additional mathematical constraints on vehicle resistance and disturbance(smooth derivable,etc.),and can realize cooperative ACC under switch topology.