"Constrained Manipulation Planning"

Dmitry Berenson, Robotics Institute, Carnegie Mellon University

November 12th (Friday), 3:00pm
ESB 2001

Manipulation planning for robotic manipulators gives rise to a rich variety of tasks that include constraints on collision-avoidance, torque, balance, closed-chain kinematics, and end-effector pose. While many researchers have developed representations and strategies to plan with a specific constraint, the goal of my work is to develop a broad representation of constraints on a robot’s configuration and identify general strategies to manage these constraints during the planning process. Some of the most important constraints in manipulation planning are functions of the pose of an arm’s end-effector, so I will devote a large part of this talk to end-effector placement for grasping and transport tasks. I will present an efficient approach to generating paths that uses Task Space Regions (TSRs) to specify manipulation tasks which involve end-effector pose goal and/or path constraints. I will show how to use TSRs for path planning with the Constrained BiDirectional RRT (CBiRRT)algorithm and describe several extensions of the TSR representation. Among them are methods to plan with object pose uncertainty and handle more complex pose constraints by chaining TSRs together. I will also show humanoid robots and mobile manipulators using CBiRRT to accomplishing practical tasks.

About Dmitry Berenson:

Dmitry Berenson is a fifth-year PhD student at Carnegie Mellon University's Robotics Institute. Before coming to CMU, he received a B.S. in Electrical and Computer Engineering from Cornell University in 2005.

His research focuses on motion and manipulation planning for robotic manipulators and humanoid robots. He is currently working on the Personal Robotics project (sponsored by Intel, CMU, and the Quality of Life Technology Center), which aims to develop a robotic assistant for the home. He has also worked with the Digital Human Research Center in Japan on grasping and motion planning algorithms and with LAAS-CNRS in France, where he focused on planning in high-dimensional cost spaces. He was awarded the Intel PhD fellowship in 2009.

Hosted by: CCDC Students Seminar