DRAPER: Towards a Robust Robot Deployment and Reliable Evaluation for Quasi-Static Pick-and-Place Cloth-Shaping Neural Controllers
This work addresses the problem of inconsistent evaluation and deployment for robotic cloth manipulation researchers, though it is incremental in improving existing methods.
The study tackled the challenge of reliably comparing simulation-trained neural controllers for robotic cloth manipulation in real-world tasks like flattening and folding, demonstrating DRAPER's framework to address grasping errors and perception gaps, achieving robust performance across different fabrics and methods.
Comparing robotic cloth-manipulation systems in a real-world setup is challenging. The fidelity gap between simulation-trained cloth neural controllers and real-world operation hinders the reliable deployment of these methods in physical trials. Inconsistent experimental setups and hardware limitations among different approaches obstruct objective evaluations. This study demonstrates a reliable real-world comparison of different simulation-trained neural controllers on both flattening and folding tasks with different types of fabrics varying in material, size, and colour. We introduce the DRAPER framework to enable this comprehensive study, which reliably reflects the true capabilities of these neural controllers. It specifically addresses real-world grasping errors, such as misgrasping and multilayer grasping, through real-world adaptations of the simulation environment to provide data trajectories that closely reflect real-world grasping scenarios. It also employs a special set of vision processing techniques to close the simulation-to-reality gap in the perception. Furthermore, it achieves robust grasping by adopting a tweezer-extended gripper and a grasping procedure. We demonstrate DRAPER's generalisability across different deep-learning methods and robotic platforms, offering valuable insights to the cloth manipulation research community.