In this research, we explore novel methods that enable human-safe robots like Baxter and Kuka to share workspace with other robots and/or humans and perform tasks collaboratively. Our focus is on non-repetitive manipulation tasks like object-picking, cleaning, and pouring.
References:
[1] K. N. Kaipa, A. S. Kankanhalli-Nagendra, N. B. Kumbla, S. Shriyam, S. S. Thevendria-Karthic, J. A. Marvel, and S. K. Gupta (2016). Addressing perception uncertainty induced failure modes in robotic bin-picking. Robotics and Computer Integrated Manufacturing 42(1), 17-38.
[2] A. Kabir, K. N. Kaipa, J. Marvel, and S. K. Gupta. Automated planning for robotic cleaning using multiple setups and oscillatory tool motions. IEEE Transactions on Automation Science and Engineering vol.PP, no.99, pp.1-14. doi: 10.1109/TASE.2017.2665460.
[3] J. D. Langsfeld, K. N. Kaipa, and S. K. Gupta. Selection of trajectory parameters for dynamic pouring tasks based on exploitation-driven updates of local metamodels, robotica. (Accepted).
[4] C. W. Morato, K. N. Kaipa, and S. K. Gupta. Toward safe human robot collaboration by using multiple Kinects based real-time human tracking (2014). ASME Journal of Computing and Information Science in Engineering, 14(1): 011006.
Infants derive a proprioceptive sense of self through “ body babbling ”, which enables them to exploit the perception that others are ” like-me “ as a starting point for social cognition. This research uses computational and robotic modeling to investigate the feasibility of the hypothesis that self-exploration is a foundational step for social cognition. Benefits of this research is two fold: (1) Design robots that display better ranges of adaptation (2) Results from these investigations can help us find the causes and treatement options for developmental disorders like cerebral palsy and autism.
Refererences:
[1] K. N. Kaipa, J. C. Bongard, and A. N. Meltzoff (2010). Self discovery enables robot social cognition: Are you my teacher? Neural Networks, Special Issue on Social Cognition: Babies to Robots, 23 (8-9): 1113-1124.
[2] K. N. Kaipa, J. C. Bongard, and A. N. Meltzoff. Combined structure and motion extraction from visual data using evolutionary active learning. Genetic and Evolutionary Computation Conference, Montreal, Canada, June 2009, pp. 121―128.
Nature abounds in examples of swarming, a form of intelligent group behavior found in insect and animal societies.The swarm intelligence mechanisms found in the natural world can inspire synthetic algorithms that can be applied to diverse fields such as optimization, multi-agent decision making, and robotics. Glowworm swarm optimization (GSO) is a swarm intelligence algorithm introduced by Kaipa and Ghose in 2005. GSO is inspired by the behavior of glowworms and is developed for numerical optimization (computing multiple optima of multimodal functions) and swarm robotics problems. Till date, GSO has been applied by the research community to a diverse variety of problems including: Multiple source localization, contaminant boundary mapping, wireless sensor networks, clustering, knapsack, numerical integration, solving fixed point equations, solving systems of nonlinear equations, and engineering design optimization. Further, other researchers have also developed several variants of GSO in order to improve its convergence properties (Visit the google scholar page on GSO for more information).
References:
[1] K. N. Kaipa and D. Ghose. Glowworm Swarm Optimization: Theory, Algorithms, and Applications, Studies in Computational Intelligence, Vol. 698, Springer-Verlag,2017. ISBN: 978-3-319-51594-6 (Print) 978-3-319-51595-3 (Online).
[2] K. N. Kaipa and D. Ghose (2009). Glowworm swarm optimization for simultaneous capture of multiple local optima of multimodal functions. Swarm Intelligence, 3(2): 87―124.
[3] K. N. Kaipa and D. Ghose. Detection of multiple source locations using a glowworm metaphor with applications to collective robotics. IEEE Swarm Intelligence Symposium, Pasadena, California, USA, June 2005, pp. 84―91.
Animals and insects have evolved distinct morphological, biomechanical, and sensorimotor structures that enable them to navigate reliably in a wide variety of unstructured terrains. Robots with a high range of adaption can be created by copying some of these critical elements into their designs. This research aims to design and build bio-inspired robots that can be used in applications like surveillance, search and rescue, and reconnaissance in difficult terrains.
Refererences:
[1] A. Vogel, K. N. Kaipa, G. Krummel, H. A. Bruck, and Satyandra K. Gupta. Design of a compliance assisted quadrupedal amphibious robot. IEEE International Conference on Robotics and Automation (ICRA 2014), Hong Kong, China, May 31-June 7, 2014.
[2] G. Krummel, K. N. Kaipa, and S. K. Gupta. Design of a horseshoe crab inspired amphibious robot for righting in surf zones. ASME Mechanisms and Robotics Conference (IDETC/CIE 2014), Buffalo, NY, August 17—20, 2014.
[3] T. Brewer, K. N. Kaipa, and S. K. Gupta. A quadruped robot with on-boarding sensing and parameterized gait for stair climbing. 15th International Conference on Climbing and Walking Robots (CLAWAR 2012), Baltimore, Maryland, USA, July 23-26, 2012, pp. 375-382.