*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Control, Safety and Coordination of Robotic Manipulators and Beyond
Committee:
Dr. Seth Hutchinson, ECE, Chair, Advisor
Dr. Sam Coogan, ECE
Dr. Ye Zhao, ME
Dr. Ousama Khatib, Stanford
Dr. Kyriakos Vamvoudakis, AE
Dr. Magnus Egerstedt, ECE
Abstract: apunov function and control barrier functions in the operational space and solved as a Quadratic Programming (QP) problem. In order to ensure real-time computation while satisfying constraints with highly coupled dynamics, we design control barriers using higher-order ellipsoids for both the task and joint constraints. This reduces the number of constraints and provides a unified framework to meet multiple objectives while ensuring exponential convergence, where applicable, and minimum-norm control in a real-time setting. We also ensure singularity avoidance by using the manipulability index and the corresponding manipulability jacobian. We also discuss the formulation to ensure safety only while utilizing any nominal controller. We apply our framework on seven degrees of freedom Kuka iiwa LBR robotic manipulator with two task constraints, two joint constraints, and one torque constraint while ensuring that manipulator is never in any singular configuration on the drake physic simulator and hardware, and the results are compared with weighted QP optimization. We then extend the formulation to the consensus problem in operational space for the heterogeneous robotic manipulator, with each manipulator satisfying its task and joint constraint while achieving consensus. Finally, we extend the formulation to wheel inverted pendulum humanoid and verify it on simulation using the Dynamic Animation and Robotics Toolkit (DART) physics engine.
