Vision-Based Robust Path Reconstruction for Robot Control
- 20 November 2013
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 63 (4) , 826-837
Abstract
Many applications in mobile robotics require the localization of the robot with respect to the boundary of a path (e.g., a lane marker) and the reconstruction of the path in front of the robot. The former can be used as a basis for a reactive control scheme that drives the robot along the specified path and the latter is used to plan its future motion. Our solution to these problems relies on a camera mounted on the chassis and pointing headway. The relevant elements of the image grabbed from the camera (i.e., the lines delimiting the path) are robustly extracted and projected into the field-of-view of a virtual camera looking over the scene from above. This way, we obtain a top plan view that allows us to reconstruct position and bearing of the vehicle irrespective of mechanical vibrations or imperfect plane of motion. What is more, by pushing forward the virtual camera, we are able to reconstruct the path in front of the robot for some distance ahead. In this paper, we describe the main ideas underlying the approach and its implementation. The accuracy of the technique and the computational workload is evaluated through a large set of experiments.Keywords
This publication has 49 references indexed in Scilit:
- High speed robotics with low cost hardware2012
- A novel system for robust lane detection and trackingSignal Processing, 2012
- Robust lane detection based on gradient direction2011
- Probabilistic Lane Tracking in Difficult Road Scenarios Using StereovisionIEEE Transactions on Intelligent Transportation Systems, 2009
- Dynamic Calibration and Occlusion Handling Algorithms for Lane TrackingIEEE Transactions on Industrial Electronics, 2009
- A Robust Lane Detection and Verification Method for Intelligent Vehicles2009
- Robust Road Modeling and Tracking Using CondensationIEEE Transactions on Intelligent Transportation Systems, 2008
- Robust lane detection in urban environments2007
- Robust Lane Lines Detection and Quantitative Assessment2007
- GOLD: a parallel real-time stereo vision system for generic obstacle and lane detectionIEEE Transactions on Image Processing, 1998