skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Whole-arm obstacle avoidance system conceptual design

Technical Report ·
DOI:https://doi.org/10.2172/6479732· OSTI ID:6479732

Whole-arm obstacle avoidance is needed for a variety of robotic applications in the Environmental Restoration and Waste Management (ER WM) Program. Typical industrial applications of robotics involve well-defined workspaces, allowing a predetermined knowledge of collision-free paths for manipulator motion. In the unstructured or poorly defined hazardous environments of the ER WM Program, the potential for significant problems resulting from collisions between manipulators and the environment in which they are utilized is great. The conceptual design for a sensing system that will provide protection against such collisions is described herein. The whole-arm obstacle avoidance system consists of a set of sensor bracelets,'' which cover the surface area of the manipulator links to the maximum extent practical, and a host processor. The host processor accepts commands from the robot control system, controls the operation of the sensors, manipulates data received from the bracelets, and makes the data available to the manipulator control system. The bracelets consist of a subset of the sensors, associated sensor interface electronics, and a bracelet interface. Redundant communications links between the host processor and the bracelets are provided, allowing single-point failure protection. The system allows reporting of 8-bit data from up to 1000 sensors at a minimum of 50 Hz. While the initial prototype implementation of the system utilizes capacitance proximity sensor, the system concept allows multiple types of sensors. These sensors are uniquely addressable, allowing remote calibration, thresholding at the bracelet, and correlation of a sensor measurement with the associated sensor and its location on the manipulator. Variable resolution allows high-speed, single-bit sensing as well as lower-speed higher-resolution sensing, which is necessary for sensor calibration and potentially useful in control.

Research Organization:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE; USDOE, Washington, DC (United States)
DOE Contract Number:
AC05-84OR21400
OSTI ID:
6479732
Report Number(s):
ORNL/TM-12337; ON: DE93012093
Country of Publication:
United States
Language:
English