Movement Research Of The Omnidirectionnal Wheeled Platform With Dislocation Structure

The omnidirectional mobile robot with three-degree-of-freedom in the plane has high flexibility and can also be applied in crowded and narrow environment.Due to the special structure of the omnidirectional wheel,the movement process often has problems such as large motion trajectory error and low stability.The BP neural network method is used to solve the vibration phenomenon and trajectory deviation of the MY3 wheel during the movement.Taking full use of the advantages of BP neural network model,the PD controller is combined with the BP model to study the algorithm of the BP-PD trajectory tracking controller,laying a foundation for improving the motion stability control of omnidirectional mobile robots,and promoting the application of the omnidirectional mobile robot.According to the motion characteristics and trajectory equation of the mobile platform,a circular BP neural network and a sinusoidal BP neural network are established.The internal parameters of the neural network and the initial angle,different speeds and trajectories are analyzed and optimized.Based on the BP neural network model,the trajectory simulation experiment is carried out.The BP model universal adaptability is verified by inputting different trajectory equations,and the superiority is judged by comparing the trajectory standard deviation.Based on the appropriate BP neural network model,the vibration of the mobile platform can be slowed down and the trajectory accuracy can be improved.The simulation results are verified by the circular trajectory field experiment.The error between the actual motion trajectory and the ideal trajectory is compared.It is proved that the BP neural network model improves the trajectory accuracy.The acceleration sensor is used to collect the comprehensive acceleration changes in the horizontal direction and the vertical direction.By comparing the accelerations of different fitting methods,it is proved that the motion stability is improved.It is judged from the two directions of trajectory error and motion stability that the circular curve BP neural network model is more adaptive.The application of BP-PD trajectory tracking controller is studied.The equations are calculated.The feasibility of the controller is verified by sinusoidal trajectory and parabolic trajectory simulation experiments.The trajectory tracking was completed for the two models before and after the speed,which confirmed the necessity of improving the stability of motion.Finally,the influence of neural network parameters on angular acceleration is analyzed,which provides a basis for effectively avoiding slippage on mobile platforms.