Research On Energy-saving Adaptive Cruise For Multiple Driving Conditions Based On Model Predictive Control

Adaptive Cruise Control(ACC)is an advanced assisted driving system that can effectively relieve the drivers driving fatigue and has been widely used in passenger cars.The traditional ACC system mostly takes the follow-up and safety as the control objectives,but ignores the vehicle's fuel economy and riding comfort.While the existing multi-target longitudinal ACC system does not fully consider the differences in driving conditions.In order to improve the adaptability of the ACC system to different driving conditions,an energy-saving adaptive cruise control system(EACC)adapted to multiple driving conditions is proposed.The main research contents are as follows:Firstly,according to the nonlinear characteristics of vehicle longitudinal dynamics,the inverse models of the driving system and the braking system are established respectively,and the driving / braking switching logic is formulated based on the threshold switching strategy,and the feedforward feedback controller is designed.The simulation results show that the designed controller can track the expected acceleration well,which lays the foundation for the development of the control strategy.Secondly,a BP neural network(Back-Propagation Neural Network)for driving condition recognition is designed.A series of characteristic parameters are screened by the correlation coefficient method,and the input parameters of the BP neural network are determined as the maximum vehicle speed,parking time,maximum acceleration,minimum acceleration,acceleration variance and driving distance,and the output parameter is the driving conditions.After that the appropriate length of the recognition window and the number of neurons in the hidden layer of the network is been chosen.Finally,the online recognition of driving conditions is verified in Simulink.The simulation results show that the designed driving condition recognition algorithm can quickly and accurately identify the driving condition types,which provides the feasibility for the design of an adaptive cruise control algorithm that adapts to the driving condition.Then,an adaptive cruise follow-up prediction model is built.On this basis,corresponding objective functions and constraints are designed for the three driving conditions(urban,rural and highway).In urban condition,the weighting of acceleration and control variable increment is taken as the objective function and the control amount,control amount increment,vehicle spacing and speed error are taken as the constraining objects;In rural condition,the weighting of economic indicator and tracking indicator is taken as the objective function,and the constraining objects are consistent with the urban condition;In highway condition,the error of expected car spacing is taken as the objective function,and the constraints are the control amount increment,vehicle spacing and vehicle speed error.Aiming at the conflict between economy object and car following object in rural conditions,multi-objective genetic algorithm is used to determine the weight coefficients.Finally,the Carsim / Simulink joint simulation platform is built.The simulations are conducted under urban,rural and highway conditions respectively.The results show that the energy-saving adaptive cruise control system adapts to multiple driving conditions ensures driving safety under three driving conditions.Under urban conditions,compared with the traditional ACC system,the vehicle using the EACC system shows a smoother acceleration profile and lower fuel consumption;Under rural conditions,compared with the traditional ACC system,the follow-up performance and fuel consumption performance of the vehicle with EACC system are better;Under highway conditions,the follow-up and fuel economy performance of vehicles with the EACC system are basically consistent with the traditional ACC system.Oversee the above analysis,it can be seen that the EACC system can select objective functions according to the type of driving conditions to achieve targeted improvement in fuel economy while ensuring driving safety.