About this document
Icra23 MS by jiwooseo523 is a document available to read on EtoBox.
The document presents a learning-based method for generating high-quality initial trajectories for trajectory optimization (TO) in path-following problems of redundant manipulators. This method, called RL-ITG, utilizes example-guided reinforcement learning to quickly produce effective initial trajectories, improving the performance of TO significantly compared to traditional methods. Experimental results demonstrate that RL-ITG enhances optimality, efficiency, and applicability in both simulations and real-
- Author
- jiwooseo523
- Language
- EN