| The accuracy and the reliability are crucial for platforms like the public service robot,automated guided vehicle for indoor logistics sorting,intelligent wheelchair and other autonomous navigation vehicle systems.Aiming at solving the problems of low motion control accuracy and high progressive error of high-power omnidirectional mobile platform,this paper proposes a series of methods based on practical demands.An offline calibration scheme for the mechanical error of three-wheel omnidirectional mobile platform based on translation and rotation motion is proposed.Closed-loop methods are adopted to reduce the yaw error,while the attitude sensing error is reduced by Butterworth low-pass filter and drift error correction methods.Extended Kalman Filter is also utilized for attitude fusion odometry.The accuracy of motion control,the attitude sensing and the odometry of mobile platform is effectively improved,meanwhile the cumulative error is greatly reduced.The main contents of this paper are as follows:Firstly,combined with the requirements of the application systems,based on the analysis and comparison of the performance and characteristics of existing mobile platforms,the kinematic analysis,overall structure design and error analysis of the three-wheel omnidirectional mobile carrier platform are completed.Secondly,referring to UMBmark method,the mechanical error model and the offline calibration method for three-wheel omnidirectional mobile platform are proposed.Through the experiments of 4 m linear translation’s yaw-mileage proportional coefficient,the measurements of rotation accuracy and other experiment steps,the error coefficients are obtained.Based on the correction of kinematic model,the closed-loop correction of yaw error and the expert PID combined with incremental and position methods,the ±1°omnidirectional motion control is finally realized.Thirdly,in order to realize the mechanical error correction,attitude sensing,motor control and multi-task parallel processing,the software based on FreeRTOS is designed,and the motion control,attitude sensing and track reckoning are realized efficiently.Finally,in order to verify the designed three-wheel omnidirectional mobile carrier platform’s performance indicators,this paper develops the functions of autonomous exploration mapping with navigation,based on the Robot Operating System,to evaluate its performance in terms of track reckoning,mapping,location accuracy.The experimental results show that the designed three-wheel omnidirectional mobile carrier platform has high motion control accuracy and relatively low progressive error.The motion control accuracy is within ±1°,the positioning angle accuracy is within ±2.5°,and the positioning error is less than 7.5 cm.The designed platform achieves the expected design goal and can be applied in various autonomous navigation systems. |