Multi-robot Formations:Diagonal/Line Following with Explorers

Robots: Three Explorers developed at the Center for Distributed Robotics in the University of Minnesota.
Use only a single monocular camera mounted on each robot !!
No inter-communication among robots, No specialized markers !!

Related Publications:

Hyeun Jeong Min, Andrew Drenner, and Nikolaos Papanikolopoulos, “Vision-based leader-follower formations with limited information,” IEEE Int. Conf. on Robotics and Automation, 2009.

erosisquare Demo
Robot Following (Explorer). (4MB mpeg)

Moving Object Segmentation from a Moving Platform

The comparison of the covariance matrices with different features
for two images showing different robots: robot 1 in 1(a) and robot 2 in 1(b). They have the same shape, but different colors. The covariance matrices are represented in images (color maps) in (c) ~ (g) for robot 1 and (h) ~ (l) for robot 2. The features in each covariance matrix are D, F, G, P, C, and Gr, which stand for distance of each pixel from the center of a bounding box, Fourier, gradient, position, colors, and gray, respectively.

Related Publications:

Hyeun Jeong Min, Nikolaos Papanikolopoulos, Christopher E. Smith, and Vassilios Morellas, “Feature-based covariance matching for a moving target in multi-robot following," submitted to The 19th Mediterranean Conf. on Control and Automation, 2011.

Hyeun Jeong Min and Nikolaos Papanikolopoulos, “Entropy-based motion segmentation from a moving platform,” IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, 2009.

Feature covariance
Multi-robot Coverage Problem for an Unknown Number of Robots

Present a novel approach to multi-robot coverage problem for an unknown number of robots. Solutions on how to cover locations of interest in a minimum time and how to find the optimal number of robots for a given time period. Specify how to find the minimum coverage time without knowing the number of robots. Analyze the algorithmical properties and propose an optimal coverage algorithm.

Related Publications:

Hyeun Jeong Min and Nikolaos Papanikolopoulos, “The Multi-robot Coverage Problem for Optimal Coordinated Searchwith an Unknown Number of Robots,” IEEE Int. Conf. on Robotics and Automation, 2011.

coverage
Delivering and deploying robots: Saddlepack and Explorers Robot following with eROSIs
erosisquare Demo

erosisquare Demo

Robot Following (eROSI). (6MB mpeg)

 

CONFERENCE TALKS

[1] IEEE Int. Conf. on Robotics and Automation, May 9 to 13, 2011 at Shanghai, China: The Multi-robot Coverage Problem for Optimal Coordinated Search with an Unknown Number of Robots.

[2] IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, October 11 to 15, 2009 at St. Louis, MO, USA: Entropy-based motion segmentation from a moving platform.

[3] IEEE Int. Conf. on Robotics and Automation, May 12 to 17, 2009 at Kobe, Japan: Vision-based leader-follower formations with limited information.

[4] SPIE Defense, Security & Sensing, Unmanned Systems Technology XI, April 13 to 17, 2009 at Orlando, FL, USA: Vision-based effective dispersion of miniature robots by using local sensing.

[5] The 15th Mediterranean Conf. on Control and Automation, 2007.: Autonomous docking for an eROSI robot based on a vision system with points clustering.

[6] Pacific Rim International Conference on Artificial Intelligence, August 9 to 13, 2004 at Auckland, New Zealand: Creative 3D designs using interactive genetic algorithm with structured directed graph.

[7] 4th Int. Symposium on Advanced Intelligent Systems, 2003 at Jeju, Korea: Automatic creation of 3D artificial flowers with interactive evolution on evolutionary engine.

[8] IEEE Int. Symposium on Computational Intelligence in Robotics and Automation, July 16 to 20, 2003 at Kobe, Japan: Generating optimal behaviors of mobile robot using behavior network with planning capability.

INVITED TALKS

[1] Invited talk from Mathematics department in Yonsei University 2009

[2] Invited talk from Mathematics department in Yonsei University 2008

[3] Invited talk from Intelligent Cognitive Robotics Lab. in Korea Institute of Science and Technology (KIST) 2008

 

PhD Thesis

Feel free to send your comments to hjmin at cs.umn.edu

    

The views and opinions expressed in this page are strictly those of the page author.
The contents of this page have not been reviewed or approved by the University of Minnesota.