Balint Thamo, Wanming Yu
Thu 22 Jul 2021, 13:00 - 14:00
Online (Blackboard collaborate)

If you have a question about this talk, please contact: Jodie Cameron (jcamero9)

Balint Thamo

Title: A Hybrid Dual Jacobian Approach for Autonomous Control of Concentric Tube Robots in Unknown Constrained Environments

Abstract: Concentric Tube Robots (CTR) have been gaining ground in minimally-invasive robotic surgeries due to their small footprint, compliance, and high dexterity. CTRs can assure safe interaction with soft tissue, provided that precise and effective motion control is achieved. Controlling the motion of CTRs is still challenging. Commonly used model-based control approaches often employ simplified geometric/dynamic assumptions, which could be very inaccurate in the presence of unmodelled disturbances and external interaction forces.

Additionally, application of emerging data-driven algorithms in real-time control of CTRs is limited due to the fact that these controllers require considerable amount of time to let the algorithm develop enough to reach a desired accuracy and relevancy.

We present a hybrid approach to overcome the aforementioned difficulties. This hybrid solution uses the solution of a kinematic model of the robot to estimate initial values for a model-free data-driven method. The proposed algorithm combines both model-based and data-driven algorithms to provide real-time motion control of CTRs interacting with an unknown external environment.


Wanming Yu

Title: Discovery of contact-rich locomotion skills for quadruped robots

Abstract: Contact-rich locomotion skills remain challenging to discover in the robotics community. In this talk, I will present an offline trajectory optimization framework which can be used for the quick discovery of feasible open loop trajectories for contact-rich locomotion tasks. The optimized trajectories are then used as an alternative source of expert demonstration for imitation learning to obtain feedback control policies which are able to generalize in various scenarios. I will demonstrate the proposed framework by showing the simulation results of a representative contact-rich locomotion task, i.e., quadruped fall recovery.