An PSO-ML-LSTM based IMU State Estimation Approach for Manipulator Teleoperation

Renyi Zhou^1,2Yuanchong Li³Aimin Zhang¹Tie Zhang⁴Yisheng Guan²Shouyan Chen^3*

• ¹Guangzhou Automobile Group Co Ltd, Guangzhou, China
• ²Guangdong University of Technology, Guangzhou, China
• ³Guangzhou University, Guangzhou, China
• ⁴South China University of Technology, Guangzhou, China

Manipulator teleoperation can liberate human from hazardous tasks. Signal noise caused by environmental disturbances and devices inherent characteristics may limit teleoperation performance. This paper proposes an approach for IMU state estimation based on particle swarm optimization and modulated long short-term memory neural networks to mitigate the impact of IMU cumulative error on robot teleoperation performance. A motion mapping model for the human arm and a 7DOF robotic arm is first established based on global configuration parameters and a hybrid mapping method. This model is used to describe the impact of IMU cumulative error on robot teleoperation performance. Subsequently, IMU pose state estimation model is constructed using particle swarm optimization (PSO) and modulated long short-term memory (ML-LSTM) neural networks. The initial data of multiple IMUs and handling handles is used for the training of the estimation model. Finally, comparative experiments are conducted to verify performance of the proposed state estimation model. The results demonstrate that the PSO-ML-LSTM algorithm can effectively eliminate the impact of IMU cumulative errors on robot teleoperation.