We do research on agile robot locomotion for bipedal and quadrupedal robots. Two strands of research in locomotion are pursued: bio-inspired and function-centric. The bio-inspired research focuses on emulating animal-like bipedal and quadruped locomotion, including walking, trotting, jumping and running. The second one, function-centric, investigates innovative design and control approaches for locomotion that are not necessarily trying to copy any biological counterpart, such as knee-less legged robots, parallel robots, etc.

We have several robots that we use to study locomotion:

SLIDER - a novel bipedal walking robot without knees.

ResQbot - an autonomous casualty extraction robot.

Robot DE NIRO - Design Engineering's Natural Interaction Robot.

DogBot - a quadrupedal walking robot from React Robotics.

Mobile Robot Fleet - our fleet of four wheeled robots.

Lab members researching robot locomotion: Ke Wang, Roni Permana Saputra, Felix Russell.

Bipedal Walking Gait Generation

Bipedal walking gait generation for a humanoid robot subjected to external pushes. The robot recovers its stability by changing the step placement and timing.Play video
Bipedal walking gait generation for a humanoid robot subjected to external pushes. The robot recovers its stability by changing the step placement and timing.

Robot WALK-MAN at DARPA Robotics Challenge

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

Humanoid robot WABIAN-2R walking with dynamically generated gait

Humanoid robot WABIAN-2R walking with dynamically generated gait. The walking gait is dynamically generated using a hybrid gait pattern generator capable of rapid and dynamically consistent pattern regeneration.Play video
Humanoid robot WABIAN-2R walking with dynamically generated gait. The walking gait is dynamically generated using a hybrid gait pattern generator capable of rapid and dynamically consistent pattern regeneration.

Compliant humanoid robot COMAN learns to walk efficiently

Compliant humanoid robot COMAN learns to walk efficiently by varying the center-of-mass height learned by reinforcement learning. The optimized walking gait achieves 18% reduction of the energy consumption.Play video
Compliant humanoid robot COMAN learns to walk efficiently by varying the center-of-mass height learned by reinforcement learning. The optimized walking gait achieves 18% reduction of the energy consumption.

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.Play video
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.

Real-time gait generation for humanoid robots

Real-time gait generation for humanoid robots. We proposed a two-stage gait pattern generation scheme for full-size humanoid robots that considers the dynamics of the whole system throughout the process.Play video
Real-time gait generation for humanoid robots. We proposed a two-stage gait pattern generation scheme for full-size humanoid robots that considers the dynamics of the whole system throughout the process.
  • Robot
  • DE NIRO

Meet our robots!

Our research is at the intersection of Robotics and Machine Learning

Contact us

Senior Lecturer (Associate Professor)
Dyson School of Design Engineering
Address: 25 Exhibition Road, South Kensington, London, SW7 2DB
YouTube of Robot Intelligence Lab
YouTube of Petar Kormushev
LinkedIn of Petar Kormushev
Twitter of Robot Intelligence Lab
Twitter of Petar Kormushev
What can our robots do?
Dyson

We are part of