Research On Safety Distance Of Adaptive Cruise Control Algorithm Based On LQR

With the continuous development of automotive electronics,computer technology and the urgent need for automotive intelligence,the Advance Driver Assistance System(ADAS)has become a crucial part of the automotive industry.As the core of ADAS system,Adaptive Cruise Control System(ACC)has important research value in enhancing vehicle control safety,reducing environmental pollution,and improving transportation efficiency.This paper takes the adaptive cruise control system of the electric vehicles as the research object and makes the following research on the its safety and comfort.First of all,taking the adaptive cruise system of electric vehicles as the research object,the ACC layered system architecture is established.Besides,we divided the ACC system into four modules: information acquisition module,decision-making central core processor,executive mechanism and human-computer interaction interface.Next,according to the vehicle following dynamic equation,vehicle inverse-longitudinal dynamics,and the improved safe distance algorithm of the fixed time interval,the adaptive cruise longitudinal dynamics model was established,and it was verified by Carsim /Simulink co-simulation.Secondly,in response to the problem of the adaptive cruise system deviating from the lane,a lane detection network based on convolutional neural layers which can extract features automatically is proposed to segment different lane lines and evaluates the position of the vehicle.Meanwhile,the accuracy and robustness of the algorithm is verified in different popular datasets.Besides,a monocular vision ACC lane ranging algorithm based on the pinhole camera imaging principle is proposed to provide advanced warning for the deviation of the ACC vehicle from the lane.Thirdly,the linear quadratic regulator(LQR)longitudinal control system was designed.Considering about the ACC safety,comfort performance,energy efficiency and vehicle power constraints,the Simultaneous Perturbation Stochastic Approximation Algorithm(SPSA)was introduced to optimize the weight matrix of the LQR controller.Then,the improved LQR longitudinal algorithm was compared with the traditional LQR algorithm through simulation to verify its performance in tracking the safety and comfort of target vehicle.Finally,the actual vehicle experiment of adaptive cruise based on safe distance is introduced.The vehicle structure,hardware foundation,communication network and software system of the experiment are described.The lane ranging algorithm and SPSALQR longitudinal control algorithm proposed in this paper are tested and analyzed under different working conditions.The experiment proves that the proposed algorithm has better performance in the safety and comfort of adaptive cruise,and the energy efficiency has also improved to a certain extent.