2020/02/12 by Johannes A. Stork, Stork, Johannes A., Todor Stoyanov +1
Computer Science · Engineering · #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Robotics (cs.RO) #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.2002.04911
openalex publication_date 2020/02/12 · openalex created_date 2022/07/24 · openalex updated_date 2026/07/28
Creating maps is an essential task in robotics and provides the basis for\neffective planning and navigation. In this paper, we learn a compact and\ncontinuous implicit surface map of an environment from a stream of range data\nwith known poses. For this, we create and incrementally adjust an ensemble of\napproximate Gaussian process (GP) experts which are each responsible for a\ndifferent part of the map. Instead of inserting all arriving data into the GP\nmodels, we greedily trade-off between model complexity and prediction error.\nOur algorithm therefore uses less resources on areas with few geometric\nfeatures and more where the environment is rich in variety. We evaluate our\napproach on synthetic and real-world data sets and analyze sensitivity to\nparameters and measurement noise. The results show that we can learn compact\nand accurate implicit surface models under different conditions, with a\nperformance comparable to or better than that of exact GP regression with\nsubsampled data.\n