Search Results

Vehicle pose estimation is a key technology for autonomous vehicles and a prerequisite for path planning and vehicle control. Visual localization has gradually attracted extensive attention from academia and industry due to its low cost and rich semantic information. However, the incremental calculation principle of the odometry inevitably leads to the accumulation of localization error with the travel distance. To solve this problem, we propose a position correction algorithm based on lightweight landmark map, and further compensate the localization error by analyzing the error characteristics. The proposed algorithm takes the stop lines on the road as landmarks, and pairs bag-of-word vectors with the positions of the corresponding landmarks. Once landmarks in the map are encountered and successfully associated, the position of the landmarks can be exploited to effectively reduce the drift of the odometry. We also present a reliable landmark map construction method.

Many learning-based methods estimate ego-motion using visual sensors. However, visual sensors are prone to intense lighting variations and textureless scenarios. 4D radar, an emerging automotive sensor, complements visual sensors effectively due to its robustness in adverse weather and lighting conditions. This paper presents an end-to-end 4D radar-visual odometry (4DRVO) approach that combines sparse point cloud data from 4D radar with image information from cameras. Using the Feature Pyramid, Pose Warping, and Cost Volume (PWC) network architecture, we extract 4D radar point features and image features at multiple scales. We then employ a hierarchical iterative refinement approach to supervise the estimated pose. We propose a novel Cross-Modal Transformer (CMT) module to effectively fuse the 4D radar point modality, image modality, and 4D radar point-image connection modality at multiple scales, achieving cross-modal feature interaction and multi-modal feature fusion.