Contact Model Fusion for Event-Based Locomotion in Unstructured Terrains

188 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Learning quadrupedal locomotion over challenging terrain (2020, positive): “This level of robustness is beyond the reach of prior controllers for ANYmal robots [1, 26] and is comparable to the state of the art [2, 29].”
Highly Dynamic Quadruped Locomotion via Whole-Body Impulse Control and Model Predictive Control (2019, methodology): “We use a periodic phase-based gait scheduler introduced in [19] that needs only two parameters to specify a gait type; phase offset and stance period for each foot.”
Dynamic Locomotion on Slippery Ground (2019, background): “Recent publications dealing with contact estimation are not addressing the slip problem [6], [7] or do not differentiate between slip and swing phases [8].”
Dynamic Locomotion in the MIT Cheetah 3 Through Convex Model-Predictive Control (2018, methodology): “Contact detection algorithms [25] allowed the gait to be modiﬁed in real time as the feet touched down earlier or later than anticipated.”
MIT Cheetah 3: Design and Control of a Robust, Dynamic Quadruped Robot (2018, background): “Since the robot does not use any external environment sensors, a contact detection algorithm probabilistically fuses encoder measurements, estimated force, and expected gait phase to estimate the likelihood that each leg is in contact with an object [11].”