Research On Obstacle Avoidance Control Method Of Intelligent Wheelchair Based On Multi-sensor Fusion

With the increasing number of aging population in my country,the number of users of smart wheelchairs is also increasing.At present,there are many researches on autonomous obstacle avoidance technology in smart wheelchairs,but most of them are aimed at the detection of larger objects on the road,and there are few researches on the detection of road dep ressions.This paper proposes a smart wheelchair obstacle avoidance control method based on multi-sensor fusion for road detection and motion control obstacle avoidance.The main research work includes the following aspects:1)Research on Pavement Detection MethodThis article studies the obstacle avoidance methods of smart wheelchairs,in road surface detection,it is necessary to obtain accurate road surface sample point feature parameters to calculate the road surface feature parameters,so a weighted multi-sensor information fusion algorithm is introduced to fuse the sensor information of smart wheelchairs,the algorithm has the problem of insufficient weight distribution accuracy in the application of road detection,so the algorithm is optimized,the iterative process of the algorithm is improved,and the fusion stability and accuracy of the algorithm are improved;in the experimental process of road detection,the analysis of the detection results found that the weighted average method could not make t he calculation result of the road obstacle characteristic value approach the optimal value,so an adaptive weighting algorithm based on the characteristic parameter of the road obstacle was designed to make the characteristic value of the road obstacle approximate best value.Based on the integration of the above algorithms,a dual-weighted multi-sensor information fusion algorithm is obtained.The actual measurement results show that the road detection method proposed in this paper can accurately locate the location and size of road obstacles.2)Research on Obstacle Avoidance Method of Intelligent Wheelchair Based on Multi-sensor.According to the operating environment of the smart wheelchair,this paper designs a multi-sensor-based smart wheelchair obstac le avoidance method,which uses Hall encoders,angle detection sensors and other sensors to feed back the current state of the wheelchair in real time,and combines the road surface detection results to generate a motion control obstacle avoidance strategy.In this paper,the differential speed steering obstacle avoidance method is introduced,and the BP neural network is introduced to optimize the PID motor control algorithm.Aiming at the problem of the slow convergence of the BP neural network algorithm,a BP neural network control algorithm based on the adaptive adjustment of the learning rate is proposed.The results show that the motion control algorithm proposed in this paper can accurately control the motor speed to avoid obstacles.3)Hardware and software design of smart wheelchairBased on the above control method,combined with the functional requirements of the smart wheelchair obstacle avoidance control,a smart wheelchair is designed.The wheelchair design is divided into hardware design and software design.Hardware design includes hardware composition structure,module circuit design,and selection of main sensors;software design includes software frame design,main program and subprogram design of each module.