Papers
45
Total Citations
1,247
H-Index
20
About
Marc D. Killpack is a robotics researcher whose work sits at the intersection of soft robotics, model-based control, and human-robot interaction. His research has fundamentally advanced how soft and inflatable robots can be controlled with precision, addressing one of the field's most persistent challenges: the difficulty of accurately modeling the complex, nonlinear dynamics of compliant systems. Killpack's most influential contributions include pioneering the use of neural networks to learn soft robot dynamics for model predictive control (MPC), a 2018 paper that has garnered 149 citations and opened new pathways for data-driven approaches in soft robotics. His earlier work applying MPC to pneumatically actuated humanoid robots (119 citations) helped establish model-based control as a viable paradigm for inherently compliant platforms. Complementing this, his research on whole-arm tactile sensing for manipulation in cluttered environments (107 citations) demonstrated how robots can safely navigate unstructured spaces using rich contact information. Across his body of work, Killpack has consistently tackled the tension between compliance and controllability, developing methods for simultaneous position and stiffness control and integrating physics-based priors into deep learning frameworks. His research has significant implications for collaborative robotics and safe human-robot interaction, with his top papers collectively accumulating over 750 citations.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Reaching in clutter with whole-arm tactile sensing107 citations · 2013
- 4Simultaneous position and stiffness control for an inflatable soft robot71 citations · 2016
- 5
- 6
- 7
- 8
- 9
- 10Model predictive control for fast reaching in clutter45 citations · 2015