General Robot Dynamics Learning and Gen2Real
This work addresses the challenge of lowering the threshold for dynamics learning in robotics, enabling amateurs to obtain models more easily, though it appears incremental in building on existing methods like dynamics randomization.
The paper tackled the problem of acquiring robot dynamics models that can generalize across different robots without restarting the training process for each new robot, achieving this by randomizing dynamics parameters, topology configurations, and model dimensions, and demonstrating validity through simulation and experiments.
Acquiring dynamics is an essential topic in robot learning, but up-to-date methods, such as dynamics randomization, need to restart to check nominal parameters, generate simulation data, and train networks whenever they face different robots. To improve it, we novelly investigate general robot dynamics, its inverse models, and Gen2Real, which means transferring to reality. Our motivations are to build a model that learns the intrinsic dynamics of various robots and lower the threshold of dynamics learning by enabling an amateur to obtain robot models without being trapped in details. This paper achieves the "generality" by randomizing dynamics parameters, topology configurations, and model dimensions, which in sequence cover the property, the connection, and the number of robot links. A structure modified from GPT is applied to access the pre-training model of general dynamics. We also study various inverse models of dynamics to facilitate different applications. We step further to investigate a new concept, "Gen2Real", to transfer simulated, general models to physical, specific robots. Simulation and experiment results demonstrate the validity of the proposed models and method.\footnote{ These authors contribute equally.