Over the course of the last few years, I have had the opportunity to work on a number of projects with prominent researchers like Tony Stentz, Lynne Parker, Bernardine Dias, Andrew Howard etc. in the robotics field that has allowed me to develop an understanding of the current state of research in the field of mobile-robots. The resultant work has been in parts exhilarating and infuriating. I also believe that some of the problems that I had the opportunity of addressing were of a fundamental nature that have scope and application beyond the completion of their respective projects.

The TreasureHunt project, rCommerce Group, Robotics Institute
Robots and humans will dynamically engage as partners in solving complex, potentially adversarial, tasks by optimally joining their complementary capabilities. There are significant, but not unachievable, challenges that must be met to realize this vision. These challenges include robust operation across multiple environments, building capabilities that are applicable across multiple robot types, building teams of robots that improve over time, and natural interactions between humans and robots.


Click on image for more details

Sliding Autonomy, rCommerce Group, Robotics Institute
A key requirement for enabling robustness and efficiency in human-robot teams is the ability to dynamically adjust the level of autonomy to optimize the use of resources and capabilities as conditions evolve. While sliding autonomy is well defined and understood in applications where a single human is working with a single robot, it is largely unexplored when applied to teams of humans working with multiple robots.


Click on image for more details

Map-based localization, rCommerce Group, Robotics Institute
The goal of the project is to produce a probability estimate of the position of a vehicle traversing from point A to point B at any point in its path. Towards that we make use of the s IMU and navigation ladar and registered aerial imagery of the operating environment. However, it is reasonable to assume there is high ambiguity in the available data. Additionally, the different modalities of the aerial and ground data makes it difficult to easily identify common features. In order to account for the ambiguity and to identify the relation between ground and aerial data, we make use of a particle filter that combines the vehicle dead-reckoning along with the probability for each particle in the sampling set given the observed ground data and pre-existing aerial imagery.




Metrics for quantifying system performance in intelligent multi-robot teams
Any system that has the capability to diagnose and recover from faults is considered to be a fault-tolerant system. Additionally, the quality of the incorporated fault-tolerance has a direct impact on the overall performance of the system. Hence, being able to measure the extent and usefulness of fault- tolerance exhibited by the system would provide the designer with a useful analysis tool for better understanding the system. We outline application-independent metrics to measure fault-tolerance within the context of system performance.tolerance towards system performance and identify potential methods for analyzing the obtained measures towards evalu- ating the true capability of a multi-robot system.


Click on image for more details

LeaF - A distributed Learning based fault-diagnostic framework for multi-robot teams, Distributed Intelligence Lab, Univ of Tennessee
At a high level, this research outlines a framework for developing a turn-key solution for fault diagnosis in complex teams of heterogeneous mobile robots. The key feature of the developed approach is its ability to learn useful information from encountered faults, unique or otherwise, towards a more robust system. Specifically, a fast learning-based approach is used that enables the robot team to autonomously detect and compensate for the wide variety of encountered faults.


Click on image for more details

The SDR Experience: Experiments with a Large-Scale Heterogeneous Mobile Robot Team, Distributed Intelligence Lab, Univ of Tennessee
This research was aimed at the development of autonomous behaviors for tightly-coupled cooperation in heterogeneous robot teams, specifically for the task of navigation assistance. These cooperative behaviors enable capable, sensor-rich "leader" robots to assist in the navigation of sensor-simple'') robots that have no onboard capabilities for obstacle avoidance or localization, and only minimal capabilities for kin recognition. To address this challenge, we developed cooperative behaviors for heterogeneous robots that enable the successful deployment of sensor-limited robots by assistance from more capable leader robots. These heterogeneous cooperative behaviors are quite complex, and involve the combination of several behavior components, including vision-based marker detection, autonomous teleoperation, color marker following in robot chains, laser-based localization, map-based path planning, and ad hoc mobile networking. To our knowledge, this is the most complex heterogeneous robot team cooperative task ever attempted on physical robots.


Videos of deployment
Simulation Videos : video1, video2
Deployment video : deployment.mpg
Single robot Deployment video : marker_deployment.mpg
There are lots more pictures and videos on the DI-LAB website, take a look.

Indoor localization techniques, Evolution Robotics
Worked with Dr. (s) Mario Munich and Jim Ostrowski at Evolution Robotics towards developing a Kalman filter system for sensor-fusion and noise elimination from a proprietary infrared based positioning device for fast and stable indoor localization. Other responsibilities included performing quality analysis for Evolution robotics SDK, ERSP (ver 2.0), specifically working on evaluating the performance of the vision-based mapping and localization software.
Home