ROVING MANIPULATOR Edith Taylor William Whittaker JSC AR&SD Field Robotics Center Objectives Manipulation and rover technologies central to HEDS are now maturing in NASA's robotics programs, and so our focus turns to combining these technologies into systems with the ability to work, assist Astronauts, and conduct science across rough terrain. Geology and planetary science depends to a great extent on micro-features ~0.1 mm and sub-surface sampling (even 1 cm) that adds a great deal to scientific knowledge. Mobiles are incompetent to deliver such resolutions, where a manipulator enables both this position and orientation of sensors and reduces rover micronavigation. Likewise, manipulators are unable to reach and service all locations within even a single HEDS construction site, where a rover's range of motion enables multiple tasks with one robot, and extends the robot's range for exploration and support of human treks across the terrain. To meet these challenges we propose to develop a new class of mobile/manipulator robot whose scale and capability will enable exploration, search, recovery, construction, logistics and maintenance tasks needed for HEDS and terrestrial applications. Joining JSC's Robonaut manipulator technology with CMU's Nomad rover development, the team's integrated configuration will combine state of the art design and control in both domains. The resulting robot will enjoy the rugged locomotion of a rover and the dextrous manipulation of a human scale arm, yielding a system whose novel scale and capabilities will be coordinated as a single electromechanism. The confluence of the Robonaut's manipulators designed for the ISS space environment with the rugged rovers demonstrated by CMU is intended to yield a space worthy candidate for moon and mars flights. JSC and CMU will team in this development with near term objectives in Antarctica, supporting a search for meteorites on that continent's ice. Similar to HEDS, the Meteorite mission requires both rugged locomotion and dextrous manipulation, with additional manipulation-on-the-fly tasks for sensor pointing and fine positioning that couples the rover and arm kinematics, dynamics and control. This system's design, integration, and coordinated control are the project's three year objectives, with multiple field tests in planetary analogs. Approach One of the key technical challenges will be the integrated design of a rugged rover and dexterous manipulator into a single, synergetic mechanism that shares common structure, power and other onboard services. JSC's work in body-arm configuration design will be coordinated with CMU research in rover optimization to make the design trades necessary for this integration. Going beyond the stop and reach mode, this system will have combined control of the base and arm as a redundant mechanism, where the arm's fine motion reduces the rover micro-navigation duties, and the rover expands the arm's workspace over planetary terrain. Coupling human scale arms with a rugged rover will enable an expanding sequence of subtasks: locate, point, touch, breakout, grasp, orient, place, contain, stow. This sequence of tasks is indicative of the Meteorite search application, and meets capability targets for HEDS. The team's major activities will be the combined rover-arm design, on-the-fly manipulator control, and field testing. The manipulator being built for the Robonaut at JSC would be used as the baseline for the arm technologies, including kinematics, actuation, sensing, servo avionics, Cartesian controller and control software. This dual use allows the Roving Manipulator to build upon an existing TRIWG project, reducing development cost and providing additional performance data to support future flights of these arms. The thermal design spec for Antarctica (-40C to 0C) is within the EVA Touch Range (-50C to 100C) that has driven the Robonaut technology development. Robonaut's design objectives of low weight (<50 lb arm), low power (<100 W), and human scale (24" reach) have yielded a manipulator that can be accommodated on a mobile platform of the scale tested by CMU without compromising rover performance. Deliverables Integrated rover-arm design, sharing onboard services and a coordinated design Ruggedized Roving Manipulator field tested in planetary analogs Fluid, coordinated control of arm and rover for on-the-fly manipulation Schedule '99 -1-2 Axis demonstration on the ice, with possible science roles (e.g. sensor pointing) -Configuration design for integrating manipulator on rover -Environmental testing of Robonaut technologies '00 -Arm integrated with rover -Roving Manipulator deployed on ice '01 -Extended field test demonstrating arm & rover coordinated control -Integration with new end-effectors -Integration with new sensors Resources Required The development of a major new class of robot will be reduced in cost by its use of and coordination with ongoing TRIWG products. The arm development will apply mechanisms, electronics and control software already in development for Robonaut, and the project's start occurs at an early enough stage in CMU's rover design to accommodate integration. The needed resources over and beyond these existing programs are 600K per year, with 500K going to JSC for arm development and 100K to CMU for integration and design. Dual Use Potential Within NASA, and TRIWG in particular, the further development of Robonaut and Nomad technologies will harden these designs to merit future ISS, lunar and mars flight opportunities. Beyond NASA's purview, the development of a ruggedized rover with human scale manipulators has a promising future for many field applications, including robotic cleanup of toxic sites, construction, agriculture, search missions, rescue, explosive mine recovery, and other various military scenarios. Partnerships JSC and CMU will team on this project, JSC leading arm design and CMU developing the rover.