Add to Chrome
Add to Firefox
Add to EdgeMultisensory Omni-directional Long-term Place Recognition: Benchmark Dataset and Analysis
Apr 18, 2017
Recognizing a previously visited place, also known as place recognition (or loop closure detection) is the key towards fully autonomous mobile robots and self-driving vehicle navigation. Augmented with various Simultaneous Localization and Mapping techniques (SLAM), loop closure detection allows for incremental pose correction and can bolster efficient and accurate map creation. However, repeated and similar scenes (perceptual aliasing) and long term appearance changes (e.g. weather variations) are major challenges for current place recognition algorithms. We introduce a new dataset Multisensory Omnidirectional Long-term Place recognition (MOLP) comprising omnidirectional intensity and disparity images. This dataset presents many of the challenges faced by outdoor mobile robots and current place recognition algorithms. Using MOLP dataset, we formulate the place recognition problem as a regularized sparse convex optimization problem. We conclude that information extracted from intensity image is superior to disparity image in isolating discriminative features for successful long term place recognition. Furthermore, when these discriminative features are extracted from an omnidirectional vision sensor, a robust bidirectional loop closure detection approach is established, allowing mobile robots to close the loop, regardless of the difference in the direction when revisiting a place.
