Pseudo Inputs Optimisation for Efficient Gaussian Process Distance Fields

Robots reason about the environment through dedicated representations. Despite the fact that Gaussian Process (GP)-based representations are appealing due to their probabilistic and continuous nature, the cubic computational complexity is a concern. In this paper, we present a novel efficient GP-based representation that has the ability to produce accurate distance fields and is parameterised by the optimal locations of pseudo inputs. When applying the proposed method together with a kernel approximation approach, we show it outperforms well-established sparse GP frameworks in efficiency and accuracy. Moreover, we extend the proposed method to work in a dynamic setting, where a map is built iteratively and the scene dynamics are accounted for by adding or removing objects from the environment representation. In a nutshell, our method provides the ability to infer dynamic distance fields and achieve state-of-the-art reconstruction efficiently.

Authors

• Lan Wu
• Cedric Le Gentil
• Teresa A. Vidal-Calleja

Keywords

• Uncertainty
• Optimization methods
• Gaussian processes
• Probabilistic logic
• Computational efficiency
• Kernel
• Computational complexity
• Gaussian Process
• Distance Map
• Pseudo Input
• Dynamic Field
• Kernel Estimation
• Representation Of The Environment
• Accurate Distance
• Root Mean Square Error
• Kernel Function
• Point Cloud
• Simulated Datasets
• Kriging
• Set Of Observations
• Local Space
• Test Points
• Covariance Function
• Inference Accuracy
• Noisy Measurements
• Input Point
• Sparse Estimation
• Dense Reconstruction
• Optimal Input
• Raw Observations
• Incremental Distance
• Sparse Point
• Exponential Kernel
• Sensor Readings
• Signed Distance Function
• Ray Casting
• Euclidean Distance Fields
• Mapping