Piyush Khandelwal

About

Piyush Khandelwal is a robotics and artificial intelligence researcher whose work bridges the gap between theoretical AI planning and practical human-robot interaction. His research spans task planning, knowledge representation, multi-robot systems, and decision-making under uncertainty — areas in which he has made substantial contributions to both academic understanding and real-world deployment. Khandelwal is perhaps best known for his role in developing the BWIBots platform, a multi-robot system designed for service tasks in open environments, which has garnered 115 citations and become a reference point for researchers working at the intersection of AI and robotics. His comparative study of PDDL- and ASP-based task planning systems (67 citations) provides practitioners with invaluable empirical guidance when selecting planners for complex domains. His work on Answer Set Programming and the action language BC for mobile robot planning demonstrates a sophisticated approach to reasoning about indirect action effects and incomplete information. Beyond planning, Khandelwal has explored probabilistic reasoning through dynamically constructed MDPs and POMDPs, and contributed to multi-robot human guidance systems. A member of the UT Austin Villa team that claimed the 2012 RoboCup Standard Platform League World Championship, his work consistently reflects a commitment to building intelligent robots that operate meaningfully alongside people.

Research Focus

Key Achievements

Top Papers

1. 1
   BWIBots: A platform for bridging the gap between AI and human–robot interaction research

   115 citations · 2017
2. 2
   Task planning in robotics: an empirical comparison of PDDL- and ASP-based systems

   67 citations · 2019
3. 3
   Planning in Action Language BC while Learning Action Costs for Mobile Robots

   31 citations · 2014
4. 4
   Dynamically Constructed (PO)MDPs for Adaptive Robot Planning

   29 citations · 2017
5. 5
   Planning in answer set programming while learning action costs for mobile robots

   20 citations · 2014
6. 6
   A Low Cost Ground Truth Detection System for RoboCup Using the Kinect

   14 citations · 2012
7. 7
   Mobile Robot Planning Using Action Language $${\mathcal {BC}}$$ with an Abstraction Hierarchy

   14 citations · 2015
8. 8
   Leading the Way: An Efficient Multi-robot Guidance System

   11 citations · 2015
9. 9
   UT Austin Villa 2012: Standard Platform League World Champions

   11 citations · 2013
10. 10
    Multi-Robot Human Guidance: Human Experiments and Multiple Concurrent Requests

    5 citations · 2017

Key Collaborators

Contact & Links