摘要
为提高移动机器人在复杂环境下的速度控制精度和适应能力,提出了一种基于多传感器融合信息的移动机器人速度控制方法。首先,根据多传感器非线性优化融合理论,通过最小化运动观测残差的方法来构建移动机器人运动状态优化估计模型。然后,对利用单目相机、轮式里程计及惯性测量单元(inertial measurement unit,IMU)观测移动机器人运动的方法进行介绍,并计算了各传感器对移动机器人运动的观测残差及其雅可比矩阵。最后,结合移动机器人运动状态估计信息与增量式PID(proportion integration differentiation,比例积分微分)控制策略,设计了移动机器人速度控制系统,并通过多项试验验证了该控制系统的性能。试验结果表明,所提出的移动机器人速度控制方法有效减小了速度估计误差,较基于轮式里程计信息的速度控制方法在精度与稳健性方面有较大提升。研究结果对提升移动机器人在复杂环境下的工作性能有显著意义。
In order to improve the speed control accuracy and adaptability of mobile robots under the complex environment,a speed control method of the mobile robot based on the multi-sensor fusion information is proposed.Firstly,according to the multi-sensor nonlinear optimization fusion theory,an optimized estimation model of mobile robot motion states was constructed by minimizing the residual errors of the motion observation.Then,the observation methods of the mobile robot motion by the monocular camera,the wheel odometer and the inertial measurement unit(IMU)were introduced,and the residual errors and their Jacobian matrices of the mobile robot motion observed by different sensors were calculated.Finally,the speed control system of the mobile robot was designed by combining the estimated motion state information with the incremental PID(proportion integration differentiation)control strategy,and the performance of this control system was verified through a number of tests.The test results showed that the proposed speed control method of the mobile robot effectively reduced the speed estimation error,and had a greater improvement in accuracy and robustness than the speed control method based on the wheel odometer information.The research result has significant significance for improving the performance of mobile robots in complex environments.
作者
王昕煜
平雪良
WANG Xin-yu;PING Xue-liang(School of Mechanical Engineering,Jiangnan University,Wuxi 214122,China;Jiangsu Key Laboratory ofAdvanced Food Manufacturing Equipment Technology,Wuxi 214122,China)
出处
《工程设计学报》
CSCD
北大核心
2021年第1期63-71,共9页
Chinese Journal of Engineering Design
