In this paper we examine the constraint equations developed for two general dynamic simulation systems from a new perspective. A contact space framework, similar to the operational space framework used for robotic control, is developed which allows the constraint equations for contact and collision to be easily specified. This framework also allows us to scrutinize the multiple relations that exist between the dynamic models used in simulation and the models originally developed for robotic control and analysis. This framework has been used to develop a simulator that can model complex interaction between generalized articulated mechanical systems at interactive rates. 1 Introduction In recent years, there have been many efforts to accurately simulate physical environments in both robotics and computer graphics. A physically accurate simulation can be used to obtain insight into the real-world behavior of a robotic, fabrication or other dynamic environment. Two recent simulator syste...

The autonomous execution of manipulation tasks in unstructured, dynamic environments requires the consideration of various motion constraints. Any motion performed during the manipulation task has to satisfy constraints imposed by the task itself, but also has to consider kinematic and dynamic limitations of the manipulator, avoid unpredictably moving obstacles, and observe constraints imposed by the global connectivity of the workspace. Furthermore, the unpredictability of unstructured environments requires the continuous incorporation of feedback to reliably satisfy these constraints. We present a novel feedback motion planning approach, called elastic roadmap framework, capable of satisfying all of the motion constraints that arise in autonomous mobile manipulation and their respective feedback requirements. This framework is validated with simulation experiments using a mobile manipulation platform and a stationary manipulator.

In recent years, many successful robotic manipulator designs have been introduced. However, there remains the challenge of designing a manipulator that possesses the inherent safety characteristics necessary for human-centered robotics. In this paper, we present a new actuation approach that has the requisite characteristics for inherent safety while maintaining the performance expected of modern designs. By drastically reducing the effective impedance of the manipulator while maintaining high frequency torque capability, we show that the competing design requirements of performance and safety can be successfully integrated into a single manipulation system. 1

This paper discusses intuitive and eficient ways to model and control the dynamics of highly redundant branching mechanisms using the operational space for-mulation. As the complexity of mechanisms increases, their modeling and control become increasingly difi-cult. The operational space formulation provides a natural framework for these problems since its basic structure provides dynamic decoupling among multiple tasks and posture behavior. Eficient recursive algo-rithms are presented for the computation of the oper-ational space dynamics of branching mechanisms with multiple operational points. The application of these algorithms results in a significant increase in the in-teractivity and usability of dynamic control of complex-branching mechanisms. The experimental results are presented using real-time dynamic simulation. 1

The possibility of controlling humanoid robots in free-space opens new fields of application involving freefloating behaviors. Recently, we presented a prioritized taskoriented control framework for the control of multiple motion primitives while complying with physical constraints imposed by the robot's body and environment. We adapt here this framework to the control of free-floating robots.

Recently, a new technology for stably levitating and controlling the position and orientation of a rigid body has been introduced. A unique feature is the use of Lorentz forces rather than the usual Maxwell forces as in magnetic bearings. The Lorentz force approach, which uses the force experiencedbyaconductor in a magnetic #eld, is seen to have several advantages. After an initial exploration phase and periodoffeasibility study, a number of potentially important applications are emerging. Among them are a way to provide #ne compliant motion for assembly, to provide high #delity force#torque feedback for teleoperation and virtual reality haptic interfaces, and to isolate sensitive payloads from environmental vibrational disturbances, either in spaceoronearth. In this paper we will discuss recent work intended to demonstrate the e#cacy of Lorentz levitation technology for these application areas. 1

Abstract — Recently, [1] suggested to derive tracking controllers for mechanical systems using a generalization of Gauss’ principle of least constraint. This method allows us to reformulate control problems as a special class of optimal control. We take this line of reasoning one step further and demonstrate that well-known and also several novel nonlinear robot control laws can be derived from this generic methodology. We show experimental verifications on a Sarcos Master Arm robot for some of the the derived controllers. We believe that the suggested approach offers a promising unification and simplification of nonlinear control law design for robots obeying rigid body dynamics equations, both with or without external constraints, with over-actuation or underactuation, as well as open-chain and closed-chain kinematics. Index Terms — Non-linear control, robot control, tracking control.

Conventional fixed-base controllers are shown not to perform well on mobile manipulators due to the dynamic interactions between a manipulator and its vehicle. An extended jacobian transpose control algorithm is developed to improve the performance of such manipulator systems. It is shown to perform well in the presence of modelling errors and the practical limitations imposed by the sensory information available for control in highly unstructured field environments. I. INTRODUCTION Robotic systems are being considered for a wide variety of applications outside their traditional factory uses, in such diverse tasks as remote maintenance, toxic waste cleanup, and fire-fighting [1-3]. These tasks require mobile robotic manipulators, i.e. manipulators carried by vehicles, see figure 1. Fig. 1. A Mobile Manipulator System. Unlike conventional industrial manipulators mounted on stationary bases, a mobile manipulator's motions interact dynamically with its vehicle, including its suspensio...

Abstract — Understanding the principles of motor coordination with redundant degrees of freedom still remains a challenging problem, particularly for new research in highly redundant robots like humanoids. Even after more than a decade of research, task space control with redundacy resolution still remains an incompletely understood theoretical topic, and also lacks a larger body of thorough experimental investigation on complex robotic systems. This paper presents our first steps towards the development of a working redundancy resolution algorithm which is robust against modeling errors and unforeseen disturbances arising from contact forces. To gain a better understanding of the pros and cons of different approaches to redundancy resolution, we focus on a comparative empirical evaluation. First, we review several redundancy resolution schemes at the velocity, acceleration and torque levels presented in the literature in a common notational framework and also introduce some new variants of these previous approaches. Second, we present experimental comparisons of these approaches on a seven-degree-of-freedom anthropomorphic robot arm. Surprisingly, one of our simplest algorithms empirically demonstrates the best performance, despite, from a theoretical point, the algorithm does not share the same beauty as some of the other methods. Finally, we discuss practical properties of these control algorithms, particularly in light of inevitable modeling errors of the robot dynamics. Index Terms — Redundancy resolution, Inverse kinematics, Task space control, Null space optimization, Dynamical decoupling

Many robot control problems of practical importance, including operational space control, can be reformulated as immediate reward reinforcement learning problems. However, few of the known optimization or reinforcement learning algorithms can be used in online learning control for robots, as they are either prohibitively slow, do not scale to interesting domains of complex robots, or require trying out policies generated by random search, which are infeasible for a physical system. Using a generalization of the EM-base reinforcement learning framework suggested by Dayan & Hinton, we reduce the problem of learning with immediate rewards to a reward-weighted regression problem with an adaptive, integrated reward transformation for faster convergence. The resulting algorithm is efficient, learns smoothly without dangerous jumps in solution space, and works well in applications of complex high degreeof-freedom robots. 1.