engleski

End-effector Force and Joint Torque Estimation of a 7-DoF Robotic Manipulator using Deep Learning

When a mobile robotic manipulator interacts with other robots, people or the environment in general, the end-effector forces need to be measured to assess if a task has been completed successfully. Traditionally used force or torque estimation methods are usually based on observers, which require knowledge of the robot dynamics. Contrary to this, our approach proposes two methods based on deep neural networks: robot end- effector force estimation and joint torque estimation. These methods require no knowledge of robot dynamics and are computationally effective but require a force sensor under the robot base. Several different architectures are considered for the tasks, and the best ones are identified among those tested. First, the data for training the networks were obtained in simulation. The trained networks showed reasonably good performance, especially using the LSTM architecture (with root- mean-squared-error – RMSE metric of 0.274 N for end-effector force estimation and 0.6189 Nm for joint torque estimation). Afterwards, data were collected on the real robot Franka Emika Panda and then used to train the same networks for joint torques estimation. The obtained results are slightly worse than in simulation (0.7778 Nm vs 0.6189 Nm, according to RMSE metric) but still reasonably good, showing the validity of the proposed approach

robotic manipulator ; force estimation ; deep learning ; neural networks

nije evidentirano