This paper describes algorithms for the animation of men and women performing three dynamic athletic behaviors: running, bicycling, and vaulting. We animate these behaviors using control algorithms that cause a physically realistic model to perform the desired maneuver. For example, control algorithms allow the simulated humans to maintain balance while moving their arms, to run or bicycle at a variety of speeds, and to perform a handspring vault. Algorithms for group behaviors allow a number of simulated bicyclists to ride as a group while avoiding simple patterns of obstacles. We add secondarymotion to the animations with springmass simulations of clothing driven by the rigid-body motion of the simulated human. For each simulation, we compare the computed motion to that of humans performing similar maneuvers both qualitatively through the comparison of real and simulated video images and quantitatively through the comparison of simulated and biomechanical data.

Biped robots have better mobility than conventional wheeled robots, but they tend to tip over easily. To be able to walk stably in various environments, such as on rough terrain, up and down slopes, or in regions containing obstacles, it is necessary for the robot to adapt to the ground conditions with a foot motion, and maintain its stability with a torso motion. When the ground conditions and stability constraint are satisfied, it is desirable to select a walking pattern that requires small torque and velocity of the joint actuators. In this paper, we first formulate the constraints of the foot motion parameters. By varying the values of the constraint parameters, we can produce different types of foot motion to adapt to ground conditions. We then propose a method for formulating the problem of the smooth hip motion with the largest stability margin using only two parameters, and derive the hip trajectory by iterative computation. Finally, the correlation between the actuator specifications and the walking patterns is described through simulation studies, and the effectiveness of the proposed methods is confirmed by simulation examples and experimental results.

Abstract — In this paper, we explore the capabilities of actuated models of the compass gait walker on rough terrain. We solve for the optimal high-level feedback policy to negotiate a perfectly known but qualitatively complex terrain, using a fixed low-level controller which selects a high-level action onceper-step. We also demonstrate that a one-step time horizon control strategy using the same low-level controller can provide performance which is surprisingly comparable to that of the infinite time horizon optimal policy. The model presented here uses a torque at the hip and an axially-directed impulsive toe-off applied just before each ground collision. Our results provide compelling evidence that actuated robots based on passive dynamic principles (e.g. no ankle torque) should inherently be capable of walking on significantly rough terrain. I.

Abstract — Provably asymptotically stable running gaits are developed for the five-link, four-actuator bipedal robot, RABBIT. A controller is designed so that the Poincaré return map associated with periodic running gaits can be computed on the basis of a model with impulse-effects that, perviously, had been used only for the design of walking gaits. This feedback design leads to the notion of a hybrid zero dynamics (HZD) for running, which in turn allows the existence and stability of running gaits to be determined on the basis of a scalar map. The main results are illustrated via simulations performed on models with known parameters and on models with parameter uncertainty and structural changes. Animations of the resulting running motions are available on the web. Index Terms — Bipedal robots; hybrid systems; limit cycles; underactuated; nonlinear control. I.

In this paper we describe an animation of a human platform diver. We simulated the motion of the diver using a dynamic model and a control system. The dynamic model is a 32 degree-of-freedom rigid body model with dynamic parameters similar to those reported in the literature for humans. The control system uses algorithms for balance, jumping, and twisting to initiate the dive, proportionalderivative servos to perform the aerial portion of the dive, and a state machine to sequence the actions throughout the dive. The motion of the simulated diver closely resembles video footage of dives performed by human athletes. The combination of dynamic simulation and a control system allowed us to animate the diver using high level commands. The control and simulation techniques presented in this paper may be useful for analysis of sports performance and for providing realistic motion for synthetic actors in computer animation and virtual environments. Keywords: Human Figure Animation, Simulatio...

Figure 1: Constrained walking skills. (a) Offline synthesis is used to generate physically-simulated motions for example problems. The example motions are used to develop a dynamics model that can make accurate step-to-step predictions. (b) This model can then be used by an online planner to navigate across constrained terrain. (c) A 3D physics-based character simulation plans steps to avoid stepping in crevasses. (d) A challenging terrain being navigated by the 3D model. Simulated characters in simulated worlds require simulated skills. We develop control strategies that enable physically-simulated characters to dynamically navigate environments with significant stepping constraints, such as sequences of gaps. We present a synthesis-analysis-synthesis framework for this type of problem. First, an offline optimization method is applied in order to compute example control solutions for randomly-generated example problems from the given task domain. Second, the example motions and their underlying control patterns are analyzed to build a lowdimensional step-to-step model of the dynamics. Third, this model is exploited by a planner to solve new instances of the task at interactive rates. We demonstrate real-time navigation across constrained terrain for physics-based simulations of 2D and 3D characters. Because the framework sythesizes its own example data, it can be applied to bipedal characters for which no motion data is available. 1

Abstract — This paper introduces MABEL, a new platform for the study of bipedal locomotion in robots. One of the purposes of building the mechanism is to explore a novel powertrain design that incorporates compliance, with the objective of improving the power efficiency of the robot, both in steady state operation and in responding to disturbances. A second purpose is to inspire the development of new feedback control algorithms for running on level surfaces and walking on rough terrain. A third motivation for building the robot is science and technology outreach; indeed, it is already included in tours when K-through-12 students visit the College of Engineering at the University of Michigan. MABEL is currently walking at 1.1 m/s on a level surface, and a related monopod at Carnegie Mellon is hopping well, establishing that the testbed has the potential to realize its many objectives.

This paper explores strategies for one class of difficult terrain: slippery surfaces. We evaluate several reflexive responses to a slip using a dynamicallysimulated, three-dimensional, bipedal robot. We explore two kinds of reaction strategies. One strategy continues the step in which the slip occurred. The other lifts the slipping foot and repositions the legs for another attempt. The most successful strategy positions the legs in a fixed triangular configuration on the step following a slip.

Many robot applications require legged robots to traverse rough or unmodeled terrain. This paper explores strategies that would enable legged robots to respond to two common types of surface contact error: slipping and tripping. Because of the rapid response required and the difficulty of sensing uneven terrain, we propose a set of reflexes that would permit the robot to react without modeling or analyzing the error condition in detail. These reflexive responses allow robust recovery from a variety of contact errors. We present simulation trials for single-slip tasks with varying coefficients of friction and single-trip tasks with varying obstacle heights. Keywords--- reactive control, reflexes, rough terrain, slipping, tripping, biped locomotion I. Introduction R OUGH terrain occurs not only in natural environments but also in environments that have been constructed or modified for human use. Currently, most legged robots lack the control techniques that would allow them to behave ...

Legged robots that operate in the real world are inherently subject to stochasticity in their dynamics and uncertainty about the terrain. Due to limited energy budgets and limited control authority, these “disturbances” cannot always be canceled out with high-gain feedback. Minimally-actuated walking machines subject to stochastic disturbances no longer satisfy strict conditions for limit-cycle stability; however, they can still demonstrate impressively long-living periods of continuous walking. Here, we employ tools from stochastic processes to examine the “stochastic stability” of idealized rimless-wheel and compass-gait walking on randomly generated uneven terrain. Furthermore, we employ tools from numerical stochastic optimal control to design a controller for an actuated compass gait model which maximizes a measure of stochastic stability- the mean first-passage-time- and compare its performance to a deterministic counterpart. Our results demonstrate that walking is well-characterized as a metastable process, and that the stochastic dynamics of walking should be accounted for during control design in order to improve the stability of our machines.