Get the Flash Player to see this rotator.


Robotics     UMD

Search robotics news archives

Bookmark and Share


Performance Evaluation and Benchmarking of Intelligent Robotic Systems

Main Participants

S.K. Gupta and R. Madhavan


This project is supported by NIST.


Mobile robot navigation, flexible automation, standardization


The field of robotics and automation holds immense promise as a key transformative technology to positively impact U.S. manufacturing. From traditional and well-established applications in the automotive industry to emerging applications such as material handling, palletizing, and logistics in warehouses, the use of robots can increase productivity whilst ensuring personnel safety. Sensing unstructured environments and automatically generating a sufficiently accurate world model (map) without re-engineering the operating environment is still an unsolved problem despite advances in sensor systems and computing power. To create useful man-machine collaborative systems and to provide continual situational awareness, a framework for generating accurate representations of the operational domain is imperative.

The proposed research will advance both fundamental research and development efforts in conjunction with the experimental verification and validation. It is envisaged that this research will yield significant insights into what performance evaluation and benchmarking efforts are needed to advance intelligent systems’ research and development. The objective standards and procedures resulting from this research will lead towards tangible and meaningful performance metrics for intelligent systems, in general, and robotics and automation systems, in particular.


The objectives of this project are:

• Development of a world modeling framework for unstructured manufacturing environments such that (a) it can provide continual and simultaneous estimates of mobile robot (AGV) positions and features in the operating environment, (b) it is sensor-agnostic and can work with multiple sensor modalities compensating for individual deficiencies of single sensors, and (c) it is sufficiently generic that it can be extended for world modeling in other domains and be of use to mobile robots intended to operate in those environments (e.g. urban search and rescue).

• Channeling efforts from lessons learnt towards standards-defining activities leading to the establishment of de facto standards.


For additional information please contact:
Dr. Raj Madhavan
Institute for Systems Research
University of Maryland College Park, MD 20742
Intelligent Systems Division
National Institute of Standards and Technology, Gaithersburg, MD 20899-8230

Phone: 301-975-2865

Back to top      
Clark School Home UMD Home ISR Home