No abstract

In search applications, autonomous unmanned vehicles must be able to efficiently reacquire and localize mobile targets that can remain out of view for long periods of time in large spaces. As such, all available information sources must be actively leveraged -including imprecise but readily available semantic observations provided by humans. To achieve this, this work develops and validates a novel collaborative human-machine sensing solution for dynamic target search. Our approach uses continuous partially observable Markov decision process (CPOMDP) planning to generate vehicle trajectories that optimally exploit imperfect detection data from onboard sensors, as well as semantic natural language observations that can be specifically requested from human sensors. The key innovation is a scalable hierarchical Gaussian mixture model formulation for efficiently solving CPOMDPs with semantic observations in continuous dynamic state spaces. The approach is demonstrated and validated with a real human-robot team engaged in dynamic indoor target search and capture scenarios on a custom testbed.

No abstract

Thank you to Prof. Nisar Ahmed for your guidance, insight, and feedback and, most of all, providing me with a job I could grow with for 5 years.Thank you to Orbit Logic team: Ken Center, Robert Glissmann, James Smith and Evan Sneath for their collaboration on the MinAu project and all of the great experiences had at Cherry Creek and San Diego. Thank you to Luke Morrissey for being a team player through all of the ups and downs of hardware deployment. Thank you to Jeremy Muesing for being a supportive friend, offering guidance and connecting me with the job at COHRINT. Thank you to Luke Burks for being an understanding boss and offering helpful guidance. Thank you to Jeff Venicx for inspiring in me a passion for underwater robotics and helping to connect me with the job at COHRINT. Thank you to the Naval Information Warface Center folks for their support in hardware deployment.