PHILIP LONG

PhD, ROBOTICIST


We propose a new method of generated and visualizing the amount of free space a manipulator has at any given configuration. This is achieved using the constrained polytope. We show how this metric can be maximized in an optimization based motion planner and the result allows collision free motion in a cluttered environment.


In this video we demonstrate how the constrained manipulability polytope can be used to evaluate a humanoid robot's workspace.


We propose a new method to evaluate the robot's performance that considers both the system's geometric structure and the presence of obstacles close to or in contact with the robot. In this case, we show how these polytopes can be combined together to evaluate the closed chain manipulability.


We propose a new method to evaluate the robot's performance that considers both the system's geometric structure and the presence of obstacles close to or in contact with the robot. This method reduces the manipulator's joint velocity limits by deforming the manipulability polytope to account for obstacles.




During my time at the RiVER lab I had the pleasure of working closely with Aykuy Onol (@aykutonol), whose work focuses mainly on contact implicit trajectory optimization. In this work we demonstrate how the proposed planning method can discover complex multi-contact motions with only high level user commands.


In this work, we define a generalized method to evaluate risk for a failure event, introducing the concept of severity based on robot state and operator experience and defining a framework to allow on-line risk monitoring.

Associated Publications