Full Workspace Generation of Serial-link Manipulators by Deep Learning based Jacobian Estimation
This work provides a faster and more accurate method for robotics researchers and engineers to compute manipulator workspaces, though it is incremental as it builds on prior research.
The paper tackles the problem of generating the full workspace of serial-link manipulators by introducing a deep-learning framework that approximates the manipulator Jacobian, resulting in significantly faster computation than numerical inverse kinematics and superior accuracy compared to other machine learning alternatives.
Apart from solving complicated problems that require a certain level of intelligence, fine-tuned deep neural networks can also create fast algorithms for slow, numerical tasks. In this paper, we introduce an improved version of [1]'s work, a fast, deep-learning framework capable of generating the full workspace of serial-link manipulators. The architecture consists of two neural networks: an estimation net that approximates the manipulator Jacobian, and a confidence net that measures the confidence of the approximation. We also introduce M3 (Manipulability Maps of Manipulators), a MATLAB robotics library based on [2](RTB), the datasets generated by which are used by this work. Results have shown that not only are the neural networks significantly faster than numerical inverse kinematics, it also offers superior accuracy when compared to other machine learning alternatives. Implementations of the algorithm (based on Keras[3]), including benchmark evaluation script, are available at https://github.com/liaopeiyuan/Jacobian-Estimation . The M3 Library APIs and datasets are also available at https://github.com/liaopeiyuan/M3 .