StarlETH, quadruped robot, Autonomous Systems Lab, Zurich, Switzerland

[Administrator]

Contributors:

Autonomous Systems Lab

Robotic Systems Lab

Home page - rsl.ethz.ch/robots-media/starleth.html

 

[Administrator]

StarlETH 3D trotting on treadmill with obstacles

Published on Aug 24, 2012

Trotting experiments with StarlETH on a treadmill with up to 0.7m/s (2.5km/h) over small obstacles. This quadrupedal robot (24kg, 0.2m segment lenght) is driven by 12 series elastic torque actuators and works with onboard state estimation. An external motion capture system is used to continuously adapt the the treadmill velocity.

 

[Administrator]

walking and running experiments with the quadruped robot StarlETH

Published on Jan 3, 2013

This movie summarizes some achievements with our quadrupedal robot StarlETH. This machine is driven by 12 high compliant series elastic acutators. The control strategy for these sequences is based on hierarchical task-space inverse dynamics.

 

[Administrator]

Published on Jun 5, 2013

StarlETH was demonstrated at the IEEE International Conference on Robotics and Automation (ICRA) in Karlsruhe, Germany. StarlETH is a electrically driven quadruped robot able to cope with unperceived obstacles with several centimeters in height. StarlETH is in active development at the Autonomous Systems Lab at ETH Zurich, Switzerland.

 

[Administrator]

Detection of Slippery Terrain with a Heterogeneous Team of Legged Robots

Published on Feb 15, 2014

This video shows a heterogneous team of legged robots conducting a joint locomotion and perception task. StarlETH, a large and highly capable quadruped uses the VelociRoACH as a remote probe to detect regions of slippery terrain. StarlETH localizes the VelociRoACH using internal state estimation (IMU and leg kinematics) and visual tracking (ARTag). While StarlETH is remote controlled, the position and orientation of the VelociRoACH is feedback controlled to stay at a desired position in front of StarlETH. A Support Vector Machines (SVM) is used to identify slippery regions with the VelociRoACH. The data for the SVM is based on data from both the external observation from StarlETH and the internal sensory data of VelociRoACH. The slippage classifier is able to detect slippery spots with 92% (125/135) accuracy using of only four features from the available data.

See Detection of slippery terrain with a heterogeneous team of legged robots, UC Berkeley and ETH Zurich

 

[Administrator]

Published on Sep 4, 2015

 

[Administrator]

Quadcopter with Quadruped or Quadruped with Quadcopter

Published on Sep 11, 2015

 

[Administrator]

StarlETH dog robot copes with tough terrain

Published on Oct 28, 2015

The StarlETH quadruped robot can run, walk, and climb over tough terrain autonomously, say its creators. Jim Drury saw the metal, four-legged, beast in action.

 

[Administrator]

Dynamic trotting on slopes for quadrupedal robots

Published on Oct 21, 2016

Quadrupedal locomotion on sloped terrains poses different challenges than walking in a mostly flat environment. The robot’s configuration needs to be explicitly controlled in order to avoid slipping and kinematic limits. To this end, information about the terrain’s inclination is required for carefully planning footholds, the pose of the main body, and modulation of the ground reaction forces. This is even more important for dynamic trotting, as only two support legs are available to compensate for gravity and drive a desired motion. We propose a reliable method for estimating the parameters of the terrain quadrupedal robots move on, in the face of limited perception capabilities and drifting robot pose estimates. The terrain information that we estimate, namely the pitch and roll angles of the ground plane, is exploited in an extended version of our previous model-based control approach. Our improved control framework enabled StarlETH, a medium-sized, fully autonomous, torque-controllable quadrupedal robot, to trot on slopes of up to 21 degree.

 

[Administrator]

Jump maneuvers with StarlETH

Published on Nov 1, 2016

The jump maneuvers were optimized in simulation and directly replayed on the quadruped robot StarlETH.