Intelligent Wheelchair Indoor Positioning Mapping And Autonomous Navigation Research

According to authoritative statistics,with the acceleration of the aging of the population and the proportion of mobility disorders caused by diseases,traffic accidents,etc.continue to rise,mobility disorders will cause serious life burden and psychological burden to themselves,wheelchairs as the main aids for mobility impairments are widely used,and its upgraded intelligent wheelchair as an important branch of the field of mobile robots,integrating a variety of robot-related technologies: positioning mapping,path planning,multi-sensor data fusion and multi-modal human-computer interaction.This paper mainly focuses on the positioning mapping algorithm and path planning algorithm in the intelligent wheelchair system,and the main research contents include:Firstly,the coordinate system model,pose model,odometer motion model and laser observation model of the intelligent wheelchair are constructed,and the expression method of the map is selected,and the traditional Monte Carlo positioning only uses a single sensor for the posture estimation of the intelligent wheelchair,resulting in a large positioning error.An improved Monte Carlo adaptive positioning algorithm is proposed,which fuses the odometer data measured by the encoder and the inertial navigation sensor during the wheelchair movement stage by extending the Kalman filter to improve the positioning accuracy.In addition,according to the number of particles required for positioning that cannot be adaptively adjusted by the traditional Monte Carlo positioning algorithm,the Coulbeck-Leibler divergence is introduced into particle resampling to improve the convergence speed of the particle swarm and solve the problems of particle degradation and lack of diversity.Secondly,the asymptotic optimal stochastic search tree RRT* algorithm for global path planning is studied,and an improvement is proposed for the RRT* algorithm to improve the problems of too many path redundant nodes,large curvature mutation and low security in global path planning.Borrowing the heuristic sampling idea and the gravitational field idea in the artificial potential field,the path redundancy nodes are pruned,and combined with the actual radius of the wheelchair,the path rotation curvature constraint strategy and the cubic B-spline curve are proposed to smooth the path.At the same time,the local path planning temporal elastic band algorithm is studied,and a comprehensive path planning algorithm based on the improved RRT* algorithm and the temporal elastic band TEB algorithm is proposed,and through simulation and experimental verification,the proposed method can plan a better and smooth path and perform local obstacle avoidance in the environment.Finally,based on the idea of modularization,the intelligent wheelchair system is built,the underlying hardware control module with the single-chip microcomputer as the core,the upper computer software platform with the robot ROS operating system as the core,and the underlying encoder speed measurement and speed closed-loop control are studied.Finally,the positioning mapping and navigation obstacle avoidance experiments are carried out indoors,and the experimental results show that the intelligent wheelchair can achieve positioning mapping,autonomous navigation and real-time obstacle avoidance in different indoor environments.