Real-time obstacle avoidance for manipulators and mobile robots
International Journal of Robotics Research
Estimating 3-D location parameters using dual number quaternions
CVGIP: Image Understanding
Iterative point matching for registration of free-form curves and surfaces
International Journal of Computer Vision
Artificial Intelligence - Special volume on computer vision
Wavelets for computer graphics: theory and applications
Wavelets for computer graphics: theory and applications
ACM Computing Surveys (CSUR)
Texture Classification by Wavelet Packet Signatures
IEEE Transactions on Pattern Analysis and Machine Intelligence
Error Analysis of a Real-Time Stereo System
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
On the Epipolar Geometry Between Two Images with Lens Distortion
ICPR '96 Proceedings of the 1996 International Conference on Pattern Recognition (ICPR '96) Volume I - Volume 7270
Dynamic stereo vision
Applied soft computing strategies for autonomous field robotics
Autonomous robotic systems
Temporal Range Registration for Unmanned Ground and Aerial Vehicles
Journal of Intelligent and Robotic Systems
Performance evaluation of temporal range registration for autonomous vehicle navigation
Integrated Computer-Aided Engineering - Performance Metrics for Intelligent Systems
Hi-index | 0.01 |
Given ambitious mission objectives and long delay timesbetween command-uplink/data-downlink sessions, increased autonomy isrequired for planetary rovers. Specifically, NASA‘s planned 2003and 2005 Mars rover missions must incorporate increased autonomy iftheir desired mission goals are to be realized. Increased autonomy, including autonomous path planning andnavigation to user designated goals, relies on good quality estimatesof the rover‘s state, e.g., its position and orientation relative tosome initial reference frame. The challenging terrain over which therover will necessarily traverse tends to seriously degrade adead-reckoned state estimate, given severe wheel slip and/orinteraction with obstacles. In this paper, we present the implementation of a complete rover navigationsystem. First, the system is able to adaptively construct semi-sparseterrain maps based on the current ground texture and distances topossible nearby obstacles. Second, the rover is able to matchsuccessively constructed terrain maps to obtain a vision-based stateestimate which can then be fused with wheel odometry toobtain a much improved state estimate. Finally the rover makes use ofthis state estimate to perform autonomous real-time path planning andnavigation to user designated goals. Reactive obstacle avoidanceis also implemented for roaming in an environment in the absence of auser designated goal. The system is demonstrated in soft soil and relatively dense rock fields, achieving state estimates that are significantly improved with respect todead reckoning alone (e.g., 0.38 m mean absolute error vs. 1.34 m), andsuccessfully navigating in multiple trials to user designated goals.