Gray C. Thomas

Summary of Team IHMC's virtual robotics challenge entry

This paper presents a high level overview of the work done by Team IHMC (Florida Institute for Human and Machine Cognition) to win the DARPA Virtual Robotics Challenge (VRC), held June 18-20 2013. The VRC consisted of a series of three tasks (driving a vehicle, walking over varied terrain, and manipulating a fire hose), to be completed in simulation using a model of the humanoid robot Atlas. Team IHMC was able to complete all of these challenges multiple times during the competition. The paper presents our approach, as well as a birds-eye view of the major software components and their integration.

Design of a Momentum-Based Control Framework and Application to the Humanoid Robot Atlas

This paper presents a momentum-based control framework for floating-base robots and its application to the humanoid robot “Atlas”. At the heart of the control framework lies a quadratic program that reconciles motion tasks expressed as constraints on the joint acceleration vector with the limitations due to unilateral ground contact and force-limited grasping. We elaborate on necessary adaptations required to move from simulation to real hardware and present results for walking across rough terrain, basic manipulation, and multi-contact balancing on sloped surfaces (the latter in simulation only). The presented control framework was used to secure second place in both the DARPA Robotics Challenge Trials in December 2013 and the Finals in June 2015.

Stabilizing Series-Elastic Point-Foot Bipeds Using Whole-Body Operational Space Control

Whole-body operational space controllers (WBOSCs) are versatile and well suited for simultaneously controlling motion and force behaviors, which can enable sophisticated modes of locomotion and balance. In this paper, we formulate a WBOSC for point-foot bipeds with series-elastic actuators (SEA) and experiment with it using a teen-size SEA biped robot. Our main contributions are on devising a WBOSC strategy for point-foot bipedal robots, 2) formulating planning algorithms for achieving unsupported dynamic balancing on our point-foot biped robot and testing them using a WBOSC, and 3) formulating force feedback control of the internal forces-corresponding to the subset of contact forces that do not generate robot motions-to regulate contact interactions with the complex environment. We experimentally validate the efficacy of our new whole-body control and planning strategies via balancing over a disjointed terrain and attaining dynamic balance through continuous stepping without a mechanical support.

Continuous cyclic stepping on 3D point-foot biped robots via constant time to velocity reversal

This paper presents a control scheme for ensuring that a 3D, under-actuated, point-foot biped robot remains balanced while walking. It achieves this by observing the center of mass (COM) position error relative to a reference path and re-planning a new reference trajectory to remove this error at every step. The Prismatic Inverted Pendulum Model (PIPM) is used to simplify behavioral analysis of the robot. We use phase space techniques to plan the COM trajectories and foot placement. While obtaining a stable path using this simplified model is easy, when applied to a real robot, there will usually be deviation from the expected path due to modeling inaccuracies. Although fully-actuated robots can reduce the deviation with relatively simple feedback control loops, when working with under-actuated robots, it is challenging to design such a feedback control loop. Our approach is based on continuous re-planning. By planning the path of the next step based on the observed initial error, we can find the proper landing location of each step. For each step we allocate sufficient time to avoid disturbances from the moment induced by the moving leg, which is not modeled in the PIPM. Our control scheme relies on the PIPM instead of the Linear Inverted Pendulum Model (LIPM) to enable non-planar COM motion, which is essential for rough terrain locomotion. We show simulation results that include full multi-body dynamics, friction, and ground reaction forces.

Fully omnidirectional compliance in mobile robots via drive-torque sensor feedback

In order to make unintentional physical interaction with robots safer for humans, we consider compliant control of an omnidirectional wheeled base. In this paper we present a fully holonomic mobile robot system which achieves compliant motion via force control, improving over previous pseudo-omnidirectional mobile systems by being fully omnidirectional. We explain our robots drive train, and present an experimental validation of our actuator control strategy. Using a smith predictor and a simple delay-based plant model, we demonstrate compliance and safe interaction in both the mobile system alone and as the base of a wheeled mobile manipulator style system.

Towards computationally efficient planning of dynamic multi-contact locomotion

This paper considers the problem of numerically efficient planning for legged robot locomotion, aiming towards reactive multi-contact planning as a reliability feature. We propose to decompose the problem into two parts: an extremely low dimensional kinematic search, which only adjusts a geometric path through space; and a dynamic optimization, which we focus on in this paper. This dynamic optimization also includes the selection of foot steps and hand-holds-in the special case of instantaneous foot re-location. This case is interesting because (1) it is a limiting behavior for algorithms with a foot switching cost, (2) it may have merit as a heuristic to guide search, and (3) it could act as a building block towards algorithms which do consider foot transition cost. The algorithm bears similarity both to phase space locomotion planning techniques for bipedal walking and the minimum time trajectory scaling problem for robot arms. A fundamental aspect of the algorithms efficiency is its use of linear programming with reuse of the active set of inequality constraints. Simulation results in a simplified setting are used to demonstrate the planning of agile locomotion behaviors.

Hybrid multi-contact dynamics for wedge jumping locomotion behaviors

Legged robots naturally exhibit continuous and discrete dynamics when maneuvering over level-ground and uneven terrains. In recent years, numerous studies have focused on locomotion hybrid dynamics. However, locomotion on more challenging terrains such as split wedges in Figure 1 has rarely been explored, let alone its hybrid dynamics. In this study, we specifically focus on a two-phase hybrid automaton formulation for this highly steep wedge locomotion. This automaton incorporates both multi-contact and flight single contact phase motions. To dynamically balance and jump upwards on this wedge, an aperiodic phase space planning is used for trajectory generations. Three control strategies are employed simultaneously: internal force control, linear and angular momentum control. Finally, simulation results are shown to verify our strategys effectiveness.

A method for dynamically balancing a point foot robot

In this paper we apply a general control framework, Whole-Body Operational Space Control (WBOSC), to an under-actuated point-foot biped. We use WBOSC to reproduce the behavior of the Prismatic Inverted Pendulum (PIP) as the center of mass dynamics for our point foot biped. We present and analyze a new algorithm that dynamically balances the PIP model by choosing footstep placements. Our algorithm uses the shooting method and numerical integration to find a footstep location even when the robots constraints do not permit an analytic solution to the model dynamics. This approach stabilizes the robot in simulation, and keeps the physical robot upright for as many as 15 steps. The primary limitation appears to be inaccurate foot positioning.

Analyzing achievable stiffness control bounds of robotic hands with coupled finger joints

The mechanical design of robotic hands has been converging towards low-inertia, tendon-driven strategies. As tendon driven robotic fingers are serial chain systems, routing strategies with compliant tendons lead to multi-articular coupling between the degrees of freedom. We propose a generalized analysis of such serial chain linkages with coupled passive joint stiffnesses. We analyze the effect of such coupling on maximum achievable stiffness control boundaries while maintaining passivity at the actuators by analytically deriving the boundaries. We believe that we can use this information to form mechanical design guidelines for intelligently selecting arrangements of compliance elements (mechanical springs) and transmission strategies, i.e. tendon routing and pulley radii, to provide intrinsic stability and customizable controller stiffness limits for high performance manipulation in robotic hands.

MIMO identification of frequency-domain unreliability in SEAs

We investigate the use of frequency domain identification and convex optimization for obtaining robust models of series elastic actuators. This early work focuses on identifying a lower bound on the ℋ∞ uncertainty, based on the non-linear behavior of the plant when identified under different conditions. An antagonistic testing apparatus allows the identification of the full two input, two output system. The aim of this work is to find a model which explains all the observed test results, despite physical non-linearity. The approach guarantees that a robust model includes all previously measured behaviors, and thus predicts the stability of never-before-tested controllers. We statistically validate the hypothesis that a single linear model cannot adequately explain the tightly clustered experimental results. And we also develop an optimization problem which finds a lower bound on the ℋ∞ uncertainty component of the robust models which we use to represent the plant in all the tested conditions.