Research On Intelligent Wheelchair Autonomous Navigation Technology In Specific Scenarios

In some specific scenarios,such as stations and airports,accompanying personnel may not be able to take care of both physically disabled wheelchair users and heavy luggage simultaneously,resulting in travel delays and losses.In other specific scenarios,such as competitions and exhibition venues,staff may not be able to look after and guide multiple physically disabled wheelchair users at the same time,and often resort to increasing labor costs as a solution.In these scenarios,when robots with autonomous navigation functions are used to assist physically disabled wheelchair users,their path planning functions usually do not take into account the need to go through accessible passages to reach their destinations.Aiming at the problems faced by physically disabled people driving wheelchairs in the above specific scenarios,this essay uses autonomous navigation technology,uses a self-developed wheelchair platform as the research object,builds an autonomous navigation system,designs the follow companion and follow path modes under autonomous navigation,and conducts experimental verification of the two modes.The main research contents and conclusions are as follows:(1)An autonomous navigation system for specific scenarios is built,and the hardware design,selection,and software function design and main controller flow design in the system are completed.(2)The follow companion mode is designed,the follow companion model is established,the UWB positioning technology is used to determine the relative position between the companion and the wheelchair,and a "inner-outer" double-layer structure follow companion controller with compensation mechanism is proposed.The outer layer uses the "distance-velocity" model to dynamically adjust the wheelchair speed,and the inner layer uses a fuzzy controller to fuse multi-source information to calculate angular velocity.In case of signal loss,the compensation mechanism enables the wheelchair to continue to maintain the following state,achieving stable following of the wheelchair.(3)The follow path mode is designed,using the Kalman filter algorithm to fuse UWB and IMU two-dimensional plane pose information;addressing the low-speed moving demand in scenarios,realizing specific scenario grid map preprocessing;based on the preprocessed grid map,proposing a segment-based shortest path planning strategy,constructing a grid map to complete simulation experiments.The results show that the segment-based shortest path planning algorithm based on the preprocessed grid map can complete path planning in specific scenarios,and improve path planning efficiency compared with the traditional A* algorithm;establishing a follow path controller to achieve the following path of the wheelchair in specific scenarios.(4)Various experiments are designed to verify the functionality and safety of the control system for following companions and following paths.The results show that under any mode,wheelchairs can meet the use requirements in specific scenarios.In follow-companion mode,wheelchairs can achieve obstacle avoidance while following in daily traffic environments and maintain stable following within a 2～2.5 m range when the following speed is ≤1.3 m/s.They can continue to maintain a following status even after losing the follow-up signal.When the speed is≤1.6 m/s,wheelchairs can achieve an emergency stop to avoid collisions,with a safety distance of ≥20 cm.In follow-path mode,the x-axis MAE is 0.169 m,the y-axis MAE is 0.143 m,and the heading angle radian MAE is 0.113 rad.The lateral error following the straight path is within 0.253 m.The control systems for both following companions and following paths can meet the driving needs of physically disabled people using wheelchairs in specific scenarios,improve traffic efficiency and safety,and enable autonomous navigation walking in specific scenarios.