This project describes a stereo image-based visual servoing (VS) system for trajectory tracking by a nonholonomic robot without externally derived pose information nor a known visual map of the environment. It is called trajectory servoing (TS). The critical component is a feature-based, indirect simultaneous localization and mapping (SLAM) method to provide a pool of available features with estimated depth, so that they may be propagated forward in time to generate image feature trajectories for VS. Short and long distance experiments show the benefits of TS for navigating unknown areas without absolute positioning. Empirically, TS has better trajectory tracking performance than pose-based feedback when both rely on the same underlying SLAM system.
Left: TS system has two major components. One (red) steers the robot to track short paths, while the other (blue) ensures the sufficiency of features to use by querying a V-SLAM module. The entire system is used when tracking long distance trajectories. Solid arrows indicate high frequency data passing, and dashed arrows low frequency. All blue arrows represent the information flow related to long distance addition. Middle: TS process uses matches from S∗ to S to define the control u, where S∗(t) is computed from the desired trajectory. Right: Feature replenishment process. There are three segments of feature trajectories. Stars are observed point sets at corresponding time. Each circle is the start or end time of next or this segment of feature trajectory. Three feature trajectories are generated by the feature replenishment.
Real experiment top view and robot view with SLAM features. Blue box is the robot’s start pose. Red box shows the end poses region of short trajectories. The green curve is a sample trajectory to track.
References
2022
TrajServo
Image-Based Trajectory Tracking Through Unknown Environments Without Absolute Positioning
Shiyu Feng, Zixuan Wu, Yipu Zhao, and 1 more author
@article{9787773,author={Feng, Shiyu and Wu, Zixuan and Zhao, Yipu and Vela, Patricio A.},journal={IEEE/ASME Transactions on Mechatronics},title={Image-Based Trajectory Tracking Through Unknown Environments Without Absolute Positioning},year={2022},volume={27},number={4},pages={2098-2106},doi={10.1109/TMECH.2022.3175819},}
