Autoencoding a Soft Touch to Learn Grasping from On-land to Underwater
This work addresses the problem of underwater robotic grasping for environmental and ocean research, offering an incremental improvement by adapting existing methods to new conditions.
This study tackled the challenge of reliable robotic grasping in underwater environments by transferring grasping knowledge from on-land to underwater using a vision-based soft robotic finger and a Supervised Variational Autoencoder (SVAE). The results showed that the SVAE model learned latent representations transferable from land to water, presenting superior adaptation to changing environments compared to commercial force-torque sensors, with improved reliability and robustness at reduced cost.
Robots play a critical role as the physical agent of human operators in exploring the ocean. However, it remains challenging to grasp objects reliably while fully submerging under a highly pressurized aquatic environment with little visible light, mainly due to the fluidic interference on the tactile mechanics between the finger and object surfaces. This study investigates the transferability of grasping knowledge from on-land to underwater via a vision-based soft robotic finger that learns 6D forces and torques (FT) using a Supervised Variational Autoencoder (SVAE). A high-framerate camera captures the whole-body deformations while a soft robotic finger interacts with physical objects on-land and underwater. Results show that the trained SVAE model learned a series of latent representations of the soft mechanics transferrable from land to water, presenting a superior adaptation to the changing environments against commercial FT sensors. Soft, delicate, and reactive grasping enabled by tactile intelligence enhances the gripper's underwater interaction with improved reliability and robustness at a much-reduced cost, paving the path for learning-based intelligent grasping to support fundamental scientific discoveries in environmental and ocean research.