Christopher L. Lewis

Decentralized control of cooperative robotic vehicles: theory and application

Describes how decentralized control theory can be used to analyze the control of multiple cooperative robotic vehicles. Models of cooperation are discussed and related to the input/output reachability, structural observability, and controllability of the entire system. Whereas decentralized control research in the past has concentrated on using decentralized controllers to partition complex physically interconnected systems, this work uses decentralized methods to connect otherwise independent nontouching robotic vehicles so that they behave in a stable, coordinated fashion. A vector Liapunov method is used to prove stability of two examples: the controlled motion of multiple vehicles along a line and the controlled motion of multiple vehicles in formation. Also presented are three applications of this theory: controlling a formation, guarding a perimeter, and surrounding a facility.

Fault tolerant operation of kinematically redundant manipulators for locked joint failures

This paper studies the degree to which the kinematic redundancy of a manipulator may be utilized for failure tolerance. A redundant manipulator is considered to be fault tolerant with respect to a given task if it is guaranteed to be capable of performing the task after any one of its joints has failed and is locked in place. A method is developed for determining the necessary constraints which insure the failure tolerance of a kinematically redundant manipulator with respect to a given critical task. This method is based on estimating the bounding boxes enclosing the self-motion manifolds for a given set of critical task points. The intersection of these bounding boxes provides a set of artificial joint limits that may guarantee the reachability of the task points after a joint failure. An algorithm for dealing with the special case of 2-D self-motion surfaces is presented, These techniques are illustrated on a PUMA 560 that is used for a 3-D Cartesian positioning task.

Dexterity optimization of kinematically redundant manipulators in the presence of joint failures

Abstract Robotic manipulators working in remote or hazardous environments require additional measures to ensure their usability upon the failure of an actuator. This work considers failure modes that result in an immobilized joint and uses the concept of worst-case dexterity to define kinematic and dynamic fault tolerance measures for redundant manipulators. These measures are then used to specify the operating configuration which is optimal in the sense that the manipulators dexterity remains high even if one of its joints fails in a locked position. The close relationship between fault tolerance and dexterity is examined using a simple planar manipulator as an example. It is demonstrated that an inverse kinematic function which maintains a high level of fault tolerance also keeps the manipulator in well-conditioned configurations known to have desirable properties.

Cooperative Sentry Vehicles And Differential GPS Leapfrog

As part of a project for the Defense Advanced Research Projects Agency, Sandia National Laboratories’ Intelligent Systems & Robotics Center is developing and testing the feasibility of using a cooperative team of robotic sentry vehicles to guard a perimeter, perform a surround task, and travel extended distances. This paper describes our most recent activities. In particular, this paper highlights the development of a Differential Global Positioning System (DGPS) leapfrog capability that allows two or more vehicles to alternate sending DGPS corrections. Using this leapfrog technique, this paper shows that a group of autonomous vehicles can travel 22.68 kilometers with a root mean square positioning error of only 5 meters.

Trajectory generation for two robots cooperating to perform a task

This paper formulates an algorithm for trajectory generation for two robots cooperating to perform an assembly task. Treating the two robots as a single redundant system, this paper derives two Jacobian matrices which relate the joint rates of the entire system to the relative motion of the grippers with respect to one another. The main advantage of this method over previous methods is that operations may be described in the coordinates of the assembly rather than in world coordinates. To join parts using this algorithm, the desired relative position is specified and the algorithm automatically determines where the robots perform the assembly. In the process of generating the trajectory, the dexterity of the highly redundant dual-robot system is taken advantage of to satisfy a variety of secondary constraints. Secondary constraints may include obstacle and joint limit avoidance and constraints to satisfy certain absolute position and orientation goals. The algorithm typically involves two phases. First the parts to be assembled are brought from an arbitrary initial configuration to an approach configuration. This phase uses a newly derived relative distance Jacobian formulation which inherently avoids collisions. Then, a more traditional Jacobian formulation is used to join the parts.

An example of failure tolerant operation of a kinematically redundant manipulator

The high cost involved in the retrieval and repair of robotic manipulators used for remediating nuclear waste, processing hazardous chemicals, or for exploring space or the deep sea, places a premium on the reliability of the system as a whole. For such applications, kinematically redundant manipulators are inherently more reliable since the additional degrees of freedom (DOF) may compensate for a failed joint. In this work, a redundant manipulator is considered to be fault tolerant with respect to a given task if it is guaranteed to be capable of performing the task after any one of its joints has failed and is locked in place. A method is developed for insuring the failure tolerance of kinematically redundant manipulators with respect to a given critical task. Techniques are developed for analyzing the manipulators workspace to find regions which are inherently suitable for critical tasks due to their relatively high level of failure tolerance. Then, constraints are imposed on the range of motion of the manipulator to guarantee that a given task is completable regardless of which joint fails. These concepts are illustrated for a PUMA 560 that is used for a three-dimensional positioning task.<<ETX>>

Trajectory generation for cooperating robots

A formulation for online trajectory generation for two robots cooperating to perform an assembly task is derived. The two robots are treated as a single redundant system. A Jacobian is formulated that relates the joint rates of the entire system to the relative motion of one of the hands with respect to the other. The minimum norm solution of this relative Jacobian equation results in a set of joint rates which perform the cooperative task. In addition to the cooperative task, secondary goals, which include obstacle and joint limit avoidance, are specified using velocities in the null space of the relative Jacobian. This formulation also allows the robots to be controlled in parallel on independent tasks

Cooperative robotic map correlation from relative position and terrain slope measurements

A method has been developed to register navigation to map coordinates using multiple robotic vehicles. The method involves measuring the relative positions of all the vehicles and correlating that template to the terrain map to generate candidate map registration values. The root sum squares (RSS) of the differences between the measured maps counterparts are computed for all possible positions of the vehicle template on the map. A lower threshold is set on the RSS differences to establish candidate locations. A grid of points around each vehicle location is established and correlation products, based on map-fitted north-south and east-west slopes for the individual vehicles over all unique candidate pairings, are computed. Vehicles demonstrating the most negatively (i.e., minimum) correlated product sums in the cardinal directions are moved and the cycle is repeated until location convergence is achieved.

Tasking and control of a squad of robotic vehicles

Sandia National Laboratories have developed a squad of robotic vehicles as a test-bed for investigating cooperative control strategies. The squad consists of eight RATLER vehicles and a command station. The RATLERs are medium-sized all-electric vehicles containing a PC104 stack for computation, control, and sensing. Three separate RF channels are used for communications; one for video, one for command and control, and one for differential GPS corrections. Using DGPS and IR proximity sensors, the vehicles are capable of autonomously traversing fairly rough terrain. The control station is a PC running Windows NT. A GUI has been developed that allows a single operator to task and monitor all eight vehicles. To date, the following mission capabilities have been demonstrated: 1. Way-Point Navigation, 2. Formation Following, 3. Perimeter Surveillance, 4. Surround and Diversion, and 5. DGPS Leap Frog. This paper describes the system and briefly outlines each mission capability. The DGPS Leap Frog capability is discussed in more detail. This capability is unique in that it demonstrates how cooperation allows the vehicles to accurately navigate beyond the RF communication range. One vehicle stops and uses its corrected GPS position to re-initialize its receiver to become the DGPS correction station for the other vehicles. Error in position accumulates each time a new vehicle takes over the DGPS duties. The accumulation in error is accurately modeled as a random walk phenomenon. This paper demonstrates how useful accuracy can be maintained beyond the vehicles range.