Research On Intelligent Wheelchair Motion Control And Automatic Obstacle Avoidance System

At present,China has gradually entered the aging society.At the same time,due to the frequent accidents and the common existence of patients with mobility inconvenience,the demand for intelligent wheelchair is increasing.With the improvement of social intelligence,the automatic obstacle avoidance of intelligent wheelchair becomes more and more important.In order to solve this problem,this paper proposes a motion control automatic obstacle avoidance algorithm based on ultrasonic sensor,and applies it to intelligent wheelchair,which has important research and engineering significance.This paper studies from the following aspects.(1)According to the analysis of users and demand of intelligent wheelchair,the performance index of obstacle avoidance system of intelligent wheelchair is proposed.According to the mechanical structure of the intelligent wheelchair,the overall structure of the intelligent wheelchair obstacle avoidance system is designed.(2)Aiming at the problem of error in ultrasonic ranging,the error sources of ultrasonic ranging are studied,and the influence of ambient temperature and measured distance on ultrasonic ranging results is analyzed.Aiming at the error sources,the error correction algorithm of ultrasonic ranging is proposed,which can correct the measurement results under arbitrary distance and arbitrary temperature within the measurement range,and improve the accuracy of ultrasonic ranging Finally,the adaptive progressive filter is used to remove the sensor noise and outliers,which further improves the accuracy of the intelligent wheelchair to obtain the obstacle information.(3)Neural network is very helpful to improve the stability and adaptability of intelligent robot system,but the traditional neural network has low learning efficiency and slow convergence speed in its training process.To solve this problem,an automatic obstacle avoidance algorithm based on reinforcement learning Saras learning algorithm and dynamic Elman neural network is proposed to improve its learning efficiency,so as to improve the automatic obstacle avoidance performance of intelligent wheelchair.(4)In order to solve the problems of fast speed,large angle,which lead to wheelchair rollover and slow speed,small angle and unable to avoid obstacles in time,the relationship between the steering angle and speed of intelligent wheelchair is studied and analyzed,and a brushless DC motor braking control method based on intention recognition is designed.The speed control of wheelchair automatic obstacle avoidance is realized by fuzzy control theory The safety of wheelchair automatic obstacle avoidance is improved.(5)Combined with the mechanical structure of intelligent wheelchair,the automatic obstacle avoidance system of intelligent wheelchair is designed,and the hardware and software of each part are researched and designed.According to the design of the intelligent wheelchair automatic obstacle avoidance system,the motion control method and automatic obstacle avoidance algorithm of intelligent wheelchair are simulated by MATLAB.The experimental results verify the feasibility of the design,improve the automatic obstacle avoidance ability of the intelligent wheelchair,and lay the foundation for the subsequent automatic obstacle avoidance system.