Reconfigurable robots have the potential for use on a much wider range of tasks than a typical, fixed-configuration robot because they may be modified to accommodate the specific task at hand. Such reconfiguration, however, necessitates the generation of a new kinematics model to allow performance of a task in Cartesian space. Whereas previous work has typically relied on either a known kit of parts from which the robot is constructed or a numerical search technique that is not guaranteed to converge, this paper presents an analytically-based kinematic identification scheme for an all-revolute, serial-link manipulator. Experimental results are presented investigating the practical utility of the technique.

The capability to assemble structures is fundamental to the use of robotics in precursor missions in orbit and on planetary surfaces. We have performed autonomous assembly in neutral buoyancy of elements of a space truss whose mating components require positioning tolerances of the same order of magnitude as the noise in the sensor systems used for the docking. Numerous trade-offs, design decisions, and innovations were made during the development of the assembly system in order to both reduce and compensate for the sensor noise. By using relative positioning, decoupling sensing and manipulation, caching high-quality position estimates, and developing a new waypoint-completion metric, we were able to reduce sensor noise to the sub-millimeter level and autonomously assemble components with millimeter tolerances. In this paper, we discuss our approaches to the problem and report the results of a series of autonomous assembly operations.

The problem of estimating external forces exerted on a robotic manipulator with harmonic drive gearing without a force-torque sensor is considered. Manipulator dynamics together with motor current feedback is used to estimate external joint torques, which are transformed into estimated external end effector forces using knowledge of the manipulator's kinematics. Adaptive control is used to tune the parameters of the robot's modeled dynamics, while adaptive radial basis function (RBF) "neural" networks are used to learn the friction model. Compliance control is implemented on a two degree of freedom manipulator based on the force estimates. Results are compared to compliance control using a six-axis force-torque sensor mounted on the manipulator

This paper describes the development of a tendon-driven, multi-fingered dexterous robot hand for space operations. This hand can achieve cylindrical grasps while producing a holding strength necessary for grasping several crew aids, tools (CATs) and interfaces. The design is based on grasping requirements for 242 CATs and interfaces found in the extravehicular activity (EVA) Tools and Equipment Reference Book. The results of this study show that over 50% of the required grasps are cylindrical, and that it may be possible for a three-fingered hand to achieve over 90% of these grasps. Further tests are implemented to show that tendon tensions agree with the predicted kinematic model.

The use of telerobots in remote locations allows humans to extend their capabilities to distant and hazardous environments. Although safer for the operator, this extension of capabilities comes at a price. Due to communication across great distances, experiencing time delays of up to several seconds is common. Any time delay in the human/robot interaction can significantly degrade the effectiveness of operator control. This paper will looks at two common instances of human-in-the-loop command of robots: command of vehicle motion and command of a manipulator to perform an assembly task. Results from three studies are summarized to show the degradation in performance caused by varying levels of time delay.

Space and underwater vehicles with robotic arms can severely tax the capability of conventional control systems. Submersible vehicles used in neutral buoyancy simulation are subject to even greater demands since they must simulate the dynamics of spacecraft in orbit as well as function as a remotely-operated underwater vehicle. In this report, the onboard control architecture, human-machine interface, and vehicle/operator communications are described for one such vehicle in operation at the University of Maryland Neutral Buoyancy Research Facility (NBRF). The Ranger Neutral Buoyancy Vehicle (RNBV) exemplifies the high-dimensional, computationally intensive nature of the current fleet of autonomous underwater vehicles while its complement of four manipulators exceeds the capabilities of most remotely operated vehicles in service today. The sensor-based, embedded onboard control system is described, and its implementation using multiple control stations is discussed.

A new paradigm is explored for controlling free-flying robots using a robot-centered coordinate strategy. The use of "dynamic" tool vectors for controlling manipulator motion is key to the development of this new approach. Variable tool vectors allow the arms to function in multiple roles, including camera positioning and self-maneuvering of the vehicle. In addition, these concepts are embodied in a resolved-rate architecture for controlling manipulators in robo-centric operations. The ranger neutral buoyancy vehicle, currently under development at the University of Maryland Space Systems Laboratory, is used to illustrate the application of this approach.

This work describes the experimental application of impedance control to a seven joint revolute manipulator built for neutral buoyancy operation. Compensator design methodologies previously applied to a single joint are extended to the full manipulator and tested in 1 g tasks. The controller was configured to mimic three useful modes at the end-effector: accommodation control, spring-dashpot control, and inertial mode. Simulated parameter errors were negligible even for outer force loop rates as low as 40 Hz. Hard contact tasks typically required damping augmentation of 5-10 times the freespace values to maintain stable contact. Preliminary experiments indicate that this controller can be a useful tool in fulfilling future objectives in space telerobotic operations.

The authors use the concept of sliding surfaces for generating two types of tracking control laws for a free-flying robot engaged in zero-gravity assembly tasks. Suction control, developed elsewhere for controlling manipulators with stationary bases, is used to track workspace trajectories for manipulators mounted on mobile platforms. Zone control is formulated for the purpose of stationkeeping a robot maneuvering unit during payload manipulation. Experimental results are described for tests performed on an air-bearing robot tracking payload trajectories along a glass surface.>

In designing a robot control system for dual arm configurations, the control engineer is faced with two challenges: to derive the equations of motion for a given situation, and to meet certain desired control requirements (for instance, minimum energy). The former may involve closed kinematic chains, such as the case when the two arms are grasping a common object. The latter usually involves nonlinear optimization. These issues are considered in the context of transporting an inertial load using two planar three-link arms. A generalized 'reduction transformation' is applied to the dynamics to remove the singularity in the system equations. A suboptimal minimum energy method is presented to reduce a difficult 12-state, six-control nonlinear optimization to two independent, nonconflicting suboptimizations. A simulation example is provided to illustrate the degree of energy reduction possible using the optimal arm torque distribution that was developed.>