1 / 14

Robot Navigation

Robot Navigation. Avneesh Sud Vaibhav Vaish under the guidance of Dr. Subhashis Banerjee. Problem Description. The aim is to enable a robot to move around on the ground plane by visually detecting and avoiding obstacles. Stages Involved (Existing). Camera Calibration

christian-house
Download Presentation

Robot Navigation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Robot Navigation Avneesh Sud Vaibhav Vaish under the guidance of Dr. Subhashis Banerjee

  2. Problem Description The aim is to enable a robot to move around on the ground plane by visually detecting and avoiding obstacles.

  3. Stages Involved (Existing) • Camera Calibration • Edge & Corner Detection • Finding Correspondences • 3-D Reconstruction • Path Planning • Navigating the Path

  4. Camera Calibration lm = Ai [Ri|ti] M • A co-ordinate system for each camera • 4+3+3 = 10 unknowns

  5. World Point X x x’ Y Y’ l’ Z Z’ C C’ e e’ Left optical center Right optical center X X’ Right Image plane Left Image plane R,T Essential & Fundamental Matrices • F = C’-TEC-1 • x’TFx = 0 • l’ = Fx • E = [T]XR • X’TEX = 0 Base Line

  6. Corner Detection • Edges are determined • Line Map is fit on the edges • Junctions of the lines are found Toolkit used : horatio

  7. Finding Correspondences (Lines) • Candidate Lines should have similar orientation • Images of end-points are got using the epipolar constraint • An new approach considers orientation of nearby lines (Amit Garg)

  8. Line Correspondence -Results

  9. Line Correspondence -Results

  10. Reconstruction Mid-point of shortest distance

  11. Locating Obstructions • The reconstructed scene is projected on the ground plane. • Clustering is done, by deleting long edges in MST • Each cluster is bounded by its convex hull TO BE DONE Identification of ground plane. Current implementation projects onto xz plane of camera co-ordinate system.

  12. Needed for Navigation • Hand-eye calibration: to locate robot w.r.t. co-ordinate system • Visual Servoing: using visual feedback for correction in motion • Path Planning: Simple backtracking algorithm OUR TARGET To complete the above by the end of the semester

  13. New Approaches Avoid Calibration by: • Self-Calibration this was attempted by Amit Garg & Deepak Verma, without much success. • Inner Camera Invariants this will let us handle varying or unknown internal parameters of the camera.

  14. References • Three Dimensional Computer Vision O. Faugeras • The Geometry of Multiple Views Andrew Zissermann • A Versatile Camera Calibration Technique Roger Tsai (IEEE J. of Rob. & Aut., 1987) • Inner Camera Invariants & Applications S. Banerjeeet al.

More Related