The design and control of a robot are very tightly coupled. The way we design a robot determines the way it can be controlled, and vice versa. Machine learning can be applied not only to optimize the robot motion controllers, but also to evolve robot designs with a particular design objective in mind.

Jumping robot with active and passive compliance

Jumping robot with active and passive compliance. Energy efficiency is achieved by using a bungee cord for energy storage during continuous hopping. We investigate how active and passive compliance can help to absorb the shock of landing impact and protect the robot.

Kinematic-free Position Control of a Robot Arm

Kinematic-free Position Control of a Robot Arm. A novel concept for position control of a robot arm based on encoderless robot controller that does not rely on any joint angle sensing.

Haptic Exploration of Unknown Surfaces with Discontinuities

Haptic Exploration of Unknown Surfaces with Discontinuities

Robot WALK-MAN at DARPA Robotics Challenge

The WALK-MAN robot is getting ready for the DARPA Robotics Challenge 2015

Testing the impedance controller and puppeteering contol mode on Robot DE NIRO

Testing the impedance controller and puppeteering contol mode on Robot DE NIRO

Autonomous Robot Valve Turning

Autonomous Robot Valve Turning. The robot learns how to react to different disturbances during the task execution in response to the changes of the uncertainty in the valve position.