RGB-D Point Cloud Registration Based on Salient Object Detection

We propose a robust algorithm for aligning rigid, noisy, and partially overlapping red green blue-depth (RGB-D) point clouds. To address the problems of data degradation and uneven distribution, we offer three strategies to increase the robustness of the iterative closest point (ICP) algorithm. First, we introduce a salient object detection (SOD) method to extract a set of points with significant structural variation in the foreground, which can avoid the unbalanced proportion of foreground and background point sets leading to the local registration. Second, registration algorithms that rely only on structural information for alignment cannot establish the correct correspondences when faced with the point set with no significant change in structure. Therefore, a bidirectional color distance (BCD) is designed to build precise correspondence with bidirectional search and color guidance. Third, the maximum correntropy criterion (MCC) and trimmed strategy are introduced into our algorithm to handle with noise and outliers. We experimentally validate that our algorithm is more robust than previous algorithms on simulated and real-world scene data in most scenarios and achieve a satisfying 3-D reconstruction of indoor scenes.

AUV localisation: a review of passive and active techniques

Localisation, i.e. estimation of one’s position in a given environment is a crucial element of many mobile systems, manned and unmanned. Due to the high demand for autonomous exploration, patrolling and inspection services and a rapid improvement of batteries, sensors and machine learning algorithms, the quality of localisation becomes even more important for smart robotic systems. The underwater domain is a very challenging environment due to the water blocking most of the signals over short distances. Recent results in localisation techniques for underwater vehicles are summarised in two principal categories: passive techniques, which strive to provide the best estimation of the vehicle’s position (global or local) given the past and current information from sensors, and active techniques, which additionally produce guidance output that is expected to minimise the uncertainty of estimated position.