Robot Learning by Demonstration with local Gaussian process regression

118 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Benchmark for Skill Learning from Demonstration: Impact of User Experience, Task Complexity, and Start Configuration on Performance (2019, background): “orized into one of four categories based on choice of model representation: statistical approaches [2], [3], dynamical systems [4]–[7], geometric techniques [8]– [10], or probabilistic inference [11]–[14]. However, comprehensive work which evaluates these approaches based on the criteria mentioned e”
Generalized Cylinders for Learning, Reproduction, Generalization, and Refinement of Robot Skills (2017, methodology): “In follow-on work, to overcome the computational cost of GP, Schneider and Ertel used local Gaussian process regression [24].”
Towards Robust Skill Generalization: Unifying Learning from Demonstration and Motion Planning (2017, background): “Although Gaussian processes (GPs) provide a non-parametric alternative [12, 13], the computational complexity of conventional GP approaches scales cubically with the number of data points, limiting their effectiveness in trajectory-based LfD settings.”
Recent Advances in Robot Learning from Demonstration (2020, background): “As the name would suggest, time-dependent policies rely on time, potentially in addition to any other form of feedback (59, 60, 62, 65).”
Trajectory-Based Skill Learning Using Generalized Cylinders (2018, methodology): “To address this issue, in follow-on work, Schneider and Ertel (2010) used local Gaussian process regression.”