Search |  Contact |  SRI Home Do not follow this link, or your host will be blocked from this site. This is a spider trap. Do not follow this link, or your host will be blocked from this site. This is a spider trap. Do not follow this link, or your host will be blocked from this site. This is a spider trap.A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A ASRI International.  333 Ravenswood Avenue.  Menlo Park, CA 94025-3493. SRI International is a nonprofit corporation.

AIC Seminar Series

Vision-Based Control and Coordination of Unmanned Vehicles

Omid ShakerniaUC Berkeley[Home Page]

Date:  2003-02-25 at 16:00

Location:  EJ228  (Directions)

   Abstract

With the goal of developing a framework for vision-based control and coordination of intelligent unmanned vehicles, we consider the following two challenge problems: 1. Vision-Based Landing of an Unmanned Air Vehicle: We present a multiple view motion estimation algorithm based on the rank deficiency of the so called multiple view matrix. We present the design and implementation of a real-time vision system to relative motion between a UAV and a (possibly moving) landing target. Compared against an on-board inertial measurement unit, flight test results show the vision-based motion estimates are accurate to within 5cm in translation and 5 degrees in rotation. We have placed the vision system in the UAV control loop and demonstrated vision-based landing onto a stationary target, and tracking of a pitching landing deck in 2 meter waves. 2. Omnidirectional Vision-Based Formation Control: We consider the problem of leader-follower formation control of nonholonomic mobile robots using omnidirectional vision. By specifying the desired formation in the image plane, we translate the control problem into a separate visual servoing task for each follower. On the vision side, we develop a multi-frame algorithm for infinitesimal multi-body motion estimation and segmentation from omnidirectional cameras. We use the proposed algorithm to segment and estimate the relative motion of the leader in the image plane of the follower. On the control side, we use feedback linearization techniques to design a tracking controller for each mobile robot which naturally incorporate collision avoidance by exploiting the geometry of omnidirectional cameras. We present experimental results on multi-body motion estimation and segmentation of a real image sequence. We also present simulation results validating the vision-based control scheme in many formation configurations.

   Bio for Omid Shakernia

Omid Shakernia received the BS degree in 1997 and the MS degree in 1999 from UC Berkeley in the Department of Electrical Engineering and Computer Sciences. He is currently pursuing a PhD in the Intelligent Machines and Robotics Lab at UC Berkeley, and is expected to file his dissertation titled "Vision-Based Control of Unmanned Vehicles" in the summer of 2003. His research interests include computer vision-based control of mobile robots, omnidirectional vision, hybrid systems, and game theory.

   Note for Visitors to SRI

Please arrive at least 10 minutes early as you will need to sign in by following instructions by the lobby phone at Building E. SRI is located at 333 Ravenswood Avenue in Menlo Park. Visitors may park in the parking lots off Fourth Street. Detailed directions to SRI, as well as maps, are available from the Visiting AIC web page. There are two entrances to SRI International located on Ravenswood Ave. Please check the Builing E entrance signage.

SRI International
©2014 SRI International 333 Ravenswood Avenue, Menlo Park, CA 94025-3493
SRI International is an independent, nonprofit corporation. Privacy policy