Autonomous Skill Acquisition on a Mobile Manipulator

By George Dimitri Konidaris, Scott Kuindersma, R. Grupen, Andrew Barto. 2011

68 citations recorded by the source corpus.

We describe a robot system that autonomously acquires skills through interaction with its environment. The robot learns to sequence the execution of a set of innate controllers to solve a task, extracts and retains components of that solution as portable skills, and then transfers those skills to reduce the time required to learn to solve a second task.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Control and Systems Engineering

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, methodology): “Konidaris et al. (2011a, 2012) proposes an approach that constructs a skill tree from human demonstrations.”
The Option-Critic Architecture (2016, methodology): “…and Ciosek 2012), but approaches that can be used naturally with continuous state and/or action spaces have only recently started to become feasible (Konidaris et al. 2011; Niekum 2013; Mann, Mannor, and Precup 2015; Mankowitz, Mann, and Mannor 2016; Kulkarni et al. 2016; Vezhnevets et al. 2016;…”
Stochastic Neural Networks for Hierarchical Reinforcement Learning (2016, background): “There has also been work on learning skills in tasks with continuous actions (Schaal et al., 2005; Konidaris et al., 2011; Daniel et al., 2013; Ranchod et al., 2015).”
From Skills to Symbols: Learning Symbolic Representations for Abstract High-Level Planning (2018, methodology): “By contrast, our approach can be considered skill-up, where we use an existing set of skills—which could be provided by a designer, or learned by the robot autonomously (Hart, 2009; Konidaris et al., 2011; Riano & McGinnity, 2012; Kompella et al., 2017) or by demonstration (Jenkins & Matarić, 2004; ”
Learning Abstract Options (2018, background): “Discovering these temporal abstractions autonomously has been the subject of extensive research [16, 28, 17, 27, 26] with approaches that can be used in continuous state and/or action spaces only recently becoming feasible [9, 20, 15, 14, 10, 31, 3].”
Autonomously constructing hierarchical task networks for planning and human-robot collaboration (2016, background): “LfD is particularly useful due to its vast applicability, as demonstrations can be effectively provided via teleoperation [8], observation [9], and kinesthetic teaching [10] to achieve task or skill proficiency.”
Options of Interest: Temporal Abstraction with Interest Functions (2020, background): “Constructing such temporal abstractions automatically from data has also been tackled extensively, and with some success (Konidaris et al. 2011; Mann, Mannor, and Precup 2015; Kulkarni et al. 2016; Mankowitz, Mann, and Mannor 2016; Bacon, Harb, and Pre-cup 2017; Machado, Bellemare, and Bowling…”
Learning tactile skills through curious exploration (2012, background): “Examples of alternative approaches to curiosity-driven learning on simple robots or simulators can be found in the work of Oudeyer et al. (2007); Vigorito and Barto (2010); Konidaris et al. (2011); Mugan and Kuipers (2012).”
Automatic Goal Generation for Reinforcement Learning Agents (2017, background): “, 2016) or for continuous tasks where demonstrations are provided (Konidaris et al., 2011; Ranchod et al., 2015).”
ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation (2020, methodology): “[38] applies HRL to Interactive Navigation (IN) tasks, navigation problems that require the agent to interact with the environment to reach its destination, such as pressing a button to open a door (see Sec.”
Robot learning from demonstration by constructing skill trees (2012, methodology): “More details are available in Konidaris et al. (2011a).”
ReLMoGen: Integrating Motion Generation in Reinforcement Learning for Mobile Manipulation (2021, background): “Notably, Konidaris et al. [39] applies HRL to Interactive Navigation (IN) tasks, problems that require the agent to interact with the environment to reach its goal.”
A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms (2019, methodology): “This is commonly done using demonstration trajectories, though it could also be performed on trajectories generated autonomously by the robot, typically after learning [143, 202, 312].”
HRL4IN: Hierarchical Reinforcement Learning for Interactive Navigation with Mobile Manipulators (2019, background): “The work by Konidaris et al. [35] is close to us because it approaches Interactive Navigation as a hierarchical reinforcement learning problem.”
RT-1: Robotics Transformer for Real-World Control at Scale (2022, methodology): “via reinforcement-learning [32, 26, 67, 37, 65, 20, 18, 11, 69], and recently through foundation models [3, 38, 66].”