Jeff Duperret’s scientific contributions

This paper develops a three degree-of-freedom sagittal-plane hybrid dynamical systems model of a bounding quadruped. Simple within-stance controls yield a closed form expression for a family of hybrid limit cycles that represent bounding behavior over a range of user-selected fore-aft speeds as a function of the model's kinematic and dynamical parameters. Controls acting on the hybrid transitions are structured so as to achieve a cascade composition of in-place bounding driving the fore-aft degree of freedom thereby decoupling the linearized dynamics of an approximation to the stride map. Careful selection of the feedback channels used to implement these controls affords infinitesimal deadbeat stability which is relatively robust against parameter mismatch. Experiments with a physical quadruped reasonably closely match the bounding behavior predicted by the hybrid limit cycle and its stable linearized approximation.

We propose the idea of a discrete navigation problem – consisting of controlling the state of a discrete-time control system to reach a goal set while in the interim avoiding a set of obstacle states – to approximate a simplified class of transitional legged robotic tasks such as leaping which have no well established mathematical description that lends itself to synthesis. The control relation given in Theorem 1 is (assuming a task solution exists) necessary and sufficient to solve a discrete navigation problem in a minimum number of steps, and is well suited to computation when a legged system’s continuous-time within-stride controller anchors sufficiently simple stance mechanics. We demonstrate the efficacy of this control technique on a physical hopping robot affixed to a boom to reactively leap over an obstacle with a running start, controlling in continuous time during stance to exhibit a linear stance map.

This technical report provides proofs and calculations for the paper "Towards Reactive Control of Simplified Legged Robotics Maneuvers," as well as implementation notes and a discussion on robustness.

We document empirically stable bounding using an actively powered spine on the Inu quadrupedal robot, and propose a reduced-order model to capture the dynamics associated with this additional, actuated spine degree of freedom. This model is sufficiently accurate as to roughly describe the robots mass center trajectory during a bounding limit cycle, thus making it a potential option for low dimensional representations of spine actuation in steady-state legged locomotion.

For direct-drive legged robots operating in unstructured environments, workspace volume and force generation are competing, scarce resources. In this paper we demonstrate that introducing geared core actuation (i.e., proximal to rather than distal from the mass center) increases workspace volume and can provide a disproportionate amount of work-producing force to the mass center without affecting leg linkage transparency. These effects are analytically quantifiable up to modest assumptions, and are demonstrated empirically on a spined quadruped performing a leap both on level ground and from an isolated foothold (an archetypal feature of unstructured terrain).

This report highlights the capabilities demonstrated during the US Army Research Laboratory Robotics Collaborative Technology Alliance Capstone Experiment that took place during October 2014. The report succinctly presents the activities of the event and provides references for further reading on the specifics of those activities. Four capabilities were evaluated as part of distinct Integrated Research Assessments (IRA): Human Robot Interaction Modalities, Semantic Navigation and Perception, Search and Observation of Doorways, and Search and Grasping of Objects in an Indoor Environment. Five task-based assessments (TBAs) using various platforms were also conducted during this timeframe, which consisted of the following capabilities: Bracing to Reach and Grasp an Object, Detection and Climbing of Stairs, Leaping over a Span, Dynamically Feasible Motion Planning, and Terrain Aware Motion Planning. The results from the IRA and TBA events characterized performance and highlighted areas that call for continued and focused efforts in order to enable advancements in intelligence-based capabilities sufficient for the teaming of autonomous systems with Soldiers.

Researchers in a variety of fields, including aeolian science, biology, and environmental science, have already made use of stationary and mobile remote sensing equipment to increase their variety of data collection opportunities. However, due to mobility challenges, remote sensing opportunities relevant to desert environments and in particular dune fields have been limited to stationary equipment. We describe here an investigative trip to two well-studied experimental deserts in New Mexico with D-RHex, a mobile remote sensing platform oriented towards desert research. D-RHex is the latest iteration of the RHex family of robots, which are six-legged, biologically inspired, small (10kg) platforms with good mobility in a variety of rough terrains, including on inclines and over obstacles of higher than robot hip height. For more information: Kod*Lab