edit SideBar


uBot-6


uBot-6 (right) with predecessor uBot-5 (left)

uBot-6 is a toddler-sized mobile manipulator that balances dynamically on two wheels. The robot has 12 degrees of freedom (DOF): two wheels, a rotatable trunk with two 4-DOF arms, and a 1-DOF head. It is 88 cm tall and weighs about 25 kg. The design of uBot-6 is based on critical evaluation of the performance of uBot-5. The redesigned mobile manipulator improves the overall performance by adjusting arm strength and robot size. Compared to uBot-5, there has been at least a threefold increase in strength of all joints. A slightly increased size improves the workspace of the robot. But the biggest change is the addition of new forms of mobility for multi-modal dexterous mobility.


uBot-6 scooting

In its primary form of locomotion the robot is a dynamic balancer on two wheels. The wheels provide non-holonomic drive capabilities with differential steering. Dynamically stable robots are well suited to environments designed for humans where both a high center of mass and a small footprint are often required. The combined use of arms and wheels for locomotion by kuckle-walking could already be shown on uBot-5 though insufficient arm strength proved a limitation here. While uBot-6 has much stronger arms, its elbows are also equipped with small, unactuated wheels that enable additional postural configurations and corresponding forms of mobility. These wheels enable statically stable four-wheeled locomotion at various body heights. The arms are used to provide Ackermann steering. uBot-6 is capable of transitioning to and between other postural configurations. Each postural configuration provides additional ways to interact with the environment and solve tasks. At the same time they also pose additional requirements to the robot design. In order to support the needs of the new postural configuration with a near horizontal body, a new head was designed. A coupled tilting mechanism supports camera adjustments for all postural configurations while maximising the field of view of the RGB-D camera.

The robot uses the same high performance low-level motor control based on custom FPGA boards as uBot-5, but on-board computing has been upgraded to a quadcore computer running linux and ROS.


uBot-6 postural configurations and transitions



Citing uBot-6:

If you would like to cite uBot-6 in your academic publications, we suggest the following citations:

For the overall design of uBot-6 and the general concept of postural configurations for multi-modal mobility and dexterous mobile manipulation:

For path planning for multiple forms of mobility:

  • Ruiken, D., Lanighan, M., and Grupen, R., Path Planning for Dexterous Mobility. Proceedings of the 24th International Conference on Automated Planning and Scheduling (ICAPS). Portsmouth, New Hampshire, June 2014.

More details on the different uBot platforms, multi-modal mobility with related path planning, and belief space planning with the uBot-6 platform can be found in:


Images

(right click image and select view to see larger)



Research Videos



Publications using the uBot-6 robot as experimental platform

  1. Ruiken, D., Lanighan, M., and Grupen, R., Postural Modes and Control for Dexterous Mobile Manipulation: the UMass uBot Concept. In Proceedings of the 13th IEEE-RAS International Conference on Humanoid Robots. Atlanta, Georgia, October 2013. Best Paper Award Finalist.
  2. Ruiken, D., Lanighan, M., and Grupen, R., Path Planning for Dexterous Mobility. Proceedings of the 24th International Conference on Automated Planning and Scheduling (ICAPS). Portsmouth, New Hampshire, June 2014.
  3. Sen, S., and Grupen, R., Manipulation Planning Using Model-Based Belief Dynamics. In Proceedings of the 13th IEEE-RAS International Conference on Humanoid Robots. Atlanta, Georgia, October 2013.
  4. Li Yang Ku, Dirk Ruiken, Erik Learned-Miller, and Rod Grupen. Error Detection and Surprise in Stochastic Robot Actions. 2015 IEEE-RAS International Conference on Humanoid Robots. Seoul, South Korea, November 2015.
  5. Jay Ming Wong, Takeshi Takahashi, and Roderic A. Grupen. Self-Supervised Deep Visuomotor Learning from Motor Unit Feedback. Are the Skeptics Right? Limits and Potentials of Deep Learning in Robotics Workshop at Robotics Science and Systems Conference (RSS), Ann Arbor, USA, June, 2016.
  6. Dirk Ruiken, Tiffany Q. Liu, Takeshi Takahashi, and Roderic A. Grupen. Reconfigurable Tasks in Belief-Space Planning. Workshop on Integrating Multiple Knowledge Representation and Reasoning Techniques in Robotics (MIRROR-16) at IROS 2016, Daejeon, Korea, October, 2016.
  7. Kyle Hollins Wray, Dirk Ruiken, Roderic A. Grupen, and Shlomo Zilberstein. Log-Space Harmonic Function Path Planning. In Proceedings of the 2016 International Conference on Robotics and Systems (IROS), Daejeon, Korea, October, 2016.
  8. Dirk Ruiken, Jay Ming Wong, Tiffany Q. Liu, Mitchell Hebert, Takeshi Takahashi, Michael W. Lanighan, and Roderic A. Grupen. Affordance-Based Active Belief: Recognition using Visual and Manual Actions. In Proceedings of the 2016 International Conference on Robotics and Systems (IROS), Daejeon, Korea, October, 2016.
  9. Jay Ming Wong and Roderic A. Grupen. Intrinsically Motivated Multimodal Structure Learning (extended version on arXiv.org: https://arxiv.org/abs/1607.04376). In the Proceedings of the IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EPIROB), Cergy-Pontoise, Paris, France, September, 2016.
  10. Dirk Ruiken, Tiffany Q. Liu, Takeshi Takahashi, and Roderic A. Grupen. Reconfigurable Tasks in Belief-Space Planning. 2016 IEEE-RAS International Conference on Humanoid Robots, Cancun, Mexico, November, 2016.
  11. Scott M. Jordan, Dirk Ruiken, Tiffany Q. Liu, Takeshi Takahashi, Michael W. Lanighan, and Roderic A. Grupen. Summary of Experiments in Belief-Space Planning at the Laboratory for Perceptual Robotics. In Proceedings of the 2017 AAAI Spring Symposia on Interactive Multi-Sensory Object Perception for Embodied Agents, Stanford, CA, USA, March, 2017.
  12. Dirk Ruiken. Belief-Space Planning for Resourceful Manipulation and Mobility. Ph.D. Dissertation, College of Information and Computer Sciences, University of Massachusetts Amherst. May, 2017.
  13. Takeshi Takahashi, Michael W. Lanighan, and Roderic A. Grupen. Hybrid Task Planning Grounded in Belief: Constructing Physical Copies of Simple Structures. Proceedings of the 27th International Conference on Automated Planning and Scheduling (ICAPS). Pittsburgh, PA, USA, June 2017.
  14. Michael W. Lanighan, Takeshi Takahashi, and Roderic A. Grupen. Robust Simple Assembly via Hierarchical Belief Space Planning. Workshop on POMDPs in Robotics: State of The Art, Challenges, and Opportunities at Robotics: Science and Systems 2017. Boston, MA, USA, July 2017.
  15. Michael W. Lanighan, Takeshi Takahashi, and Roderic A. Grupen. Planning Robust Manual Tasks in Hierarchical Belief Spaces. Proceedings of the 28th International Conference on Automated Planning and Scheduling (ICAPS). Delft, The Netherlands, June 2018.
  16. Takeshi Takahashi, Michael W. Lanighan, and Roderic A. Grupen. Intrinsically Motivated Self-Supervised Deep Sensorimotor Learning for Grasping. In Proceedings of the 2018 International Conference on Robotics and Systems (IROS), Madrid, Spain, October, 2018.
  17. Li Yang Ku. Integration of Robotic Perception, Action, and Memory. Ph.D. Dissertation, College of Information and Computer Sciences, University of Massachusetts Amherst. May, 2018.
  18. Michael W. Lanighan. Hierarchical Belief Spaces for Autonomous Mobile Manipulation. Ph.D. Dissertation, College of Information and Computer Sciences, University of Massachusetts Amherst. May, 2019.