Tony Dear

ASSOCIATE IN DISCIPLINE

Research Interests

Robot locomotion, motion planning , geometric mechanics

Tony Dear is interested in developing a unified geometric framework for a variety of locomoting robotic systems. In particular, by using the same mathematical language to describe different robot configurations, he seeks to reduce high-dimensional complexity and overcome physical effects that may not be traditionally modeled in idealized systems.

Dear’s background in robotics is the application of geometric mechanics to locomoting systems. Geometric mechanics takes ideas from fields such as differential geometry and grounds them in explaining the underlying structure of different mechanical systems. This becomes useful for gaining intuition in motion planning, as long as the system of interest satisfies certain assumptions, for example in terms of actuation, kinematic nature, or symmetries. Because these assumptions can be restrictive, Dear has been working to combine the intuitive simplicity of geometric methods with the more computational nature of methods in learning and nonlinear control in order to apply these results to real robot systems.

Dear received his BS in Electrical Engineering and Computer Science from UC Berkeley in 2012. He subsequently received his MS in 2015 and PhD in 2018, both in Robotics from Carnegie Mellon University. He also completed the Future Faculty Program while at CMU.

RESEARCH EXPERIENCE

  • Graduate research assistant, Biorobotics Lab, Carnegie Mellon University, 2012-2018

Selected Publications

  • Tony Dear, Scott David Kelly, Matthew Travers, and Howie Choset, “Locomotion of a Multi-link Nonholonomic Snake Robot,” Proceedings of the ASME 2017 Dynamics Systems and Control Conference, Tysons Corner, October 2017.
  • Tony Dear, Scott David Kelly, and Howie Choset, “Control and Locomotion of Hydrodynamically Coupled Rigid Spheres,” Proceedings of the 2017 American Control Conference, Seattle, May 2017.
  • Ross L. Hatton, Tony Dear, and Howie Choset, “Kinematic Cartography and the Efficiency of Viscous Swimming,” IEEE Transactions on Robotics, Vol. 33, No. 3, 523-535 (2017).
  • Tony Dear, Scott David Kelly, Matthew Travers, and Howie Choset, “Locomotive Analysis of a Single-Input Three-Link Snake Robot,” Proceedings of the 55th IEEE Conference on Decision and Control, Las Vegas, December 2016.
  • Tony Dear, Scott David Kelly, and Howie Choset, “Variations on the Role of Principal Connections in Robotic Locomotion,” Proceedings of the ASME 2016 Dynamics Systems and Control Conference, Minneapolis, October 2016.
  • Tony Dear, Scott David Kelly, Matthew Travers, and Howie Choset, “The Three-Link Nonholonomic Snake as a Hybrid Kinodynamic System,” Proceedings of the 2016 American Control Conference, Boston, July 2016.