Research On Obstacle Avoidance Control Of Intelligent Vehicle Based On Pothole Detection

In recent years,with the increase of road service life,a large number of road potholes have been produced,which severely exerts an influence on people’s comfortable driving feelings.However,most of the existing active obstacle avoidance control algorithms just focus on collision avoidance of vehicles and pedestrians,and lack of avoidance of traffic accident risks caused by negative obstacles such as road potholes.To solve this serious problem,an active obstacle avoidance control algorithm is proposed on the basis of pothole detection in the paper,aiming to reduce the accident rate of vehicles driving on the pothole road.Firstly,research in recent years related to road pothole detection,driving comfort and vehicle active obstacle avoidance is introduced,the strengths and weaknesses of the existing research are also expounded,and then the research ideas and main points of this paper are specified.In addition,an algorithm for feature extraction of potholes is proposed by using multi-sensor fusion technology,including area calculation and contour extraction of road potholes based on visual images.It also includes depth measurement of potholes based on single point ranging laser radar.To solve the problem that the pothole image is easily affected by illumination and road traffic markings,the MSR multi-scale algorithm and Hough transform image preprocessing algorithm are proposed to enhance the gray characteristics of the image and eliminate the interference of road traffic markings.What’s more,the slope barrier,depth barrier of the pothole boundary and vehicle parameter characteristics are considered comprehensively and then the pass ability judgment formula of the pothole is raised,which is used to predict whether the vehicle can pass the pothole safely.Through building vehicle model and road pothole model,the influence of pothole area and depth on driving comfort is analyzed,and based on ADAMS simulation test,according to the evaluation index of human comfort and vibration brittle value of transportation products,the limit speeds of different sizes pothole are established, meeting the requirements of human comfort and the integrity of transportation goods.Then,on the basis of fuzzy control algorithm,an active obstacle avoidance control algorithm which considering dangerous road potholes is designed.The deviation between the actual distance of potholes and the safe distance,as well as the difference between the limit speed of crossing potholes and the actual speed are all considered to calculate the acceleration of the intelligent vehicle.The proposed algorithm can guarantee the vehicle passing through the pothole road cozily and securely.The CarSim/Simulink co-simulation model is built to simulate and analyze four kinds of pothole road conditions to verify the effectiveness of the algorithm.Finally,in order to further verify the control effect of the designed vehicle active obstacle avoidance control algorithm based on pothole detection,the software and hardware performance tests and real vehicle tests of the selected sensors,controllers and actuators are carried out.The results indicate that the proposed algorithm can control the vehicle passing through the pothole road safely and smoothly,and meet the requirements of passenger comfort and the integrity of the transport goods,which provides a reference for future research on the pothole road obstacle avoidance control.