Towards Robust Skill Generalization: Unifying Learning from Demonstration and Motion Planning

By Muhammad Asif Rana, Mustafa Mukadam, Seyed Reza Ahmadzadeh, Sonia Chernova, Byron Boots. 2017

93 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.

Citations in the corpus

Benchmark for Skill Learning from Demonstration: Impact of User Experience, Task Complexity, and Start Configuration on Performance (2019, background): “Based on the choice of motion generation strategy, the majority of skill learning approaches can be further broadly categorized into four groups: time-invariant (stable) dynamical systems [2], [3], [13], time-dependent dynamical systems [7], [9], [10], time-dependent statistical methods [8], [14], [”
One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning (2018, background): “acquire a range of complex skills through demonstration, such as table tennis [28], lane following [34], pouring water [31], drawer opening [38], and multi-stage manipulation tasks [62].”
Federated Imitation Learning: A Privacy Considered Imitation Learning Framework for Cloud Robotic Systems with Heterogeneous Sensor Data (2019, background): “Prior work has shown that robots can acquire a range of complex skills through demonstration, such as table tennis [1], drawer opening [2], and multi-stage manipulation tasks [3].”
Trajectory-Based Skill Learning Using Generalized Cylinders (2018, background): “CLAMP, on the other hand, generates trajectories that naturally follow the demonstrated policy while satisfying the constraints.”
Kernelized movement primitives (2017, unknown): “Other related work is Gaussian process based probabilistic inference (Dong et al. 2016; Rana et al. 2017).”
ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation (2020, background): “The opposite has also been attempted: improving the exploration of RL agents using experiences from a MG [27, 28, 29].”
Tool Macgyvering: Tool Construction Using Geometric Reasoning (2019, background): “Alternatively, this information could be learned from demonstrations [24], including, if necessary, adapting the original action to fit the dimensions of the new tool [25], [26].”
ReLMoGen: Integrating Motion Generation in Reinforcement Learning for Mobile Manipulation (2021, background): “The opposite has also been attempted: improving the exploration of RL agents using experiences from a MG [27, 28, 29, 30].”
Recent Advances in Robot Learning from Demonstration (2020, background): “Several approaches to trajectory learning, such as dynamical systems (70, 82, 96, 97, 99, 100) and probabilistic inference (13, 62, 81), have been studied in the literature.”
Differentiable Gaussian Process Motion Planning (2019, background): “These may be provided by an asymptotically optimal (but slow) motion planner [22] or by human demonstration [23]. dGPMP2 can be trained to produce similar trajectories by minimizing an error metric between the demonstrations and learner’s output with where θ e is the expert’s demonstrated…”