Backstepping Control Of Wheelchair Robot Based On CAN Communication

With the aging population and people with disabilities caused by various diseases and disasters increasing every year, researchers all around the world are carrying out a wide range of technical studies that focus on the wheelchair robot in order to provide superior performance means of transport for the old and disabled people.It is proposed a motion control based on rear wheel differential drive system of wheelchair robot which can achieve rotation of wheelchair in relatively small space of the interior on the basis of analyzing the advantages and disadvantages of the structural design of wheelchair robot at home and abroad. Taking into account balance and stability of wheelchair robot, it adds two universal wheels and completes function of support and movement control of the wheelchair. According to the requirements of perception of environment as well as the state feedback of internal motion of wheelchair robot, a modular and multi-sensor communication system is designed based on CAN (Controller Area Network) bus. Using the features of real-time and fault tolerance of CAN bus, it can obtain the information of ultrasonic sensors, laser sensors and visual sensors to achieve the goals of obstacle avoidance, positioning, target detection and recognition of wheelchair robot. Through the optical encoders installed in the driving wheels, it can get the wheelchair robot's speed of internal state information and achieve the feedback of wheelchair motion control.After the establishment of multi-sensor communication system, wheelchair robot needs to transform the data information obtained from a number of sensors into the state vectors which can express the motion state of wheelchair robot. Ultrasonic and laser sensors are the same type of sensors, which can achieve the goals of positioning the position and detecting the obstacles of wheelchair robot based on the processed information. It can get the image information according to the visual sensors and complete target recognition using image processing. And also with triangular rule, it can calculate the distance between the targets and the orientation angle that the wheelchair robot needs to deflect.Based on getting the motion status information of wheelchair robot, backstepping controllers are designed for the wheelchair robot according to the characteristics of non-linear and non-coordination of rear wheel differential drive of the wheelchair robot. According to the kinematics and kinetics equations of wheelchair robot, mathematical model of the motion state is established. The state parameters of wheelchair robot are obtained by the multi-sensor communication system. At the same time, simulation of the designed controller using Matlab/Simulink is implemented on the basis of positions and the different obstacles of wheelchair robot. Simulation results show that the designed backstepping controller can achieve error free tracking for a given trajectory and the validity and robustness of the controller are verified.