I am a PhD candidate in Electrical Engineering at the Center for Robotics and Intelligent Machines. My primary area of research is two-fold:
- Sensor network guided intelligent navigation:
The research is inspired by applications in the fields of search-&-rescue, emergency response, agriculture and military. Currently I am focusing on developing intelligent algorithms to localize targets in unknown environments without the aid of global positioning information. This allows the algorithm to be distributed and location-independent. The algorithm uses the received signal strength in randomly deployed wireless sensors to generate a pseudo-gradient in the region. The gradient peak is located at or close-to the targets in the region. The goal is to have mobile autonomous robots navigate from any point within the WSN-covered region to an identified target through interaction with the WSN. Reaching the target is achieved in two ways, by: (i) producing gradients in the WSN region that give the maximum peak closest to the target, (ii) having the sensors guide the robot to the target efficiently. This scheme uses the topology of the network to create a navigation path. The robot follows the pseudo-gradient in the WSN to reach the global maximum. The WSN and robots only make use of the relative neighborhood information to attain their goal.
- Coordinated multiple robots with sensor networks:
This research follows a three-pronged approach to the problem of WSN-assisted mobile robot navigation:
- Localizing occurring events in a region using static WSN.
- Navigating multiple mobile robots to multiple identified target locations efficiently and reliably.
- Enabling coordinated tasks to be performed by multiple mobile robots at target locations.
One of the key actions to be performed by emergent event responders is cooperative transport, whether for search-&-rescue, dousing fires or clearing objects from the pathways for easy traffic of responders. In a dynamic environment with multiple events and multiple mobile robots, it is desirable that the mobile robots form teams on-the-fly, execute a coordinated action and then disband to serve other target locations. The coordinated task is performed by the mobile robots with goal-directed movement. The coordination of the mobile robots is a result of a Virtual-Spring based interaction mechanism among the robots, which allows the robots to move together as a team. The robot team follows the pseudo-gradient in the region created by the WSN, and transports the object towards a goal.
- Distributed, Intelligent, Coordinated Robotics
- Distributed Control with Wireless Sensing
- Computer Vision and Artificial Neural Networks
- Embedded Systems and Artificial Intelligence
- Optimization Methods along with Statistical Pattern Classification
- Nikhil Deshpande, Edward Grant, and Thomas C. Henderson, "Target-directed Navigation using Wireless Sensor Networks and Implicit Surface Interpolation", In Proceedings of 2012 IEEE International Conference on Robotics and Automation, (ICRA 2012), May 14-18, 2012, St. Paul, Minnesota, USA.
- Nikhil Deshpande, Edward Grant, and Thomas C. Henderson, "Wireless Sensor Network Localization and Navigation - A Pseudo-Gradient Algorithm Approach", <submitted under review>, IEEE Systems Journal, Special Issue "Sensor Networks for Advanced Localization Systems", April 2011.
- Thomas Henderson, Hongchang Peng, Nikhil Deshpande, and Edward Grant, "The Role of Symmetry in Structural Bootstrapping", in Bridges between the Methodological and Practical Work of the Robotics and Cognitive Systems Communities - From Sensors to Concepts, Amirat Y., Chibani A., Zarri G. P. (eds.), Springer series on Intelligent Systems Reference Library, December 2011.
- Kyle Luthy, Edward Grant, Nikhil Deshpande, and Thomas C. Henderson “Perimeter Detection in Wireless Sensor Networks”, Robotics and Autonomous Systems, Volume 60, Issue 2, February 2012, Pages 266-277
- Thomas Henderson, Hongchang Peng, K. Sikorski, Nikhil Deshpande, and Edward Grant, "The Cognitive Symmetry Engine: An Active Approach to Knowledge", In Proceedings of 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011), Workshop on Knowledge Representation for Autonomous Robots, Sept. 25-30, 2011, San Francisco, California, USA.
- Nikhil Deshpande, Edward Grant, and Thomas C. Henderson, "Experiments with a “Pseudo-Gradient” Algorithm for Target Localization Using Wireless Sensor Networks", In Proceedings of 2010 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, (MFI 2010), Sept. 5-7, 2010, Salt Lake City, Utah, USA.