Reinforcement Learning in the Multi-Robot Domain

By Maja J. Mataric. 1997

Subfields: Reinforcement Learning, Robotics & Control

How researchers see this paper

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

In Reinforcement Learning

In Robotics & Control

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, background): “Other examples include foraging (Matarić, 1997) and cooperative tasks (Matarić, 1994) with multiple robots, grasping with restricted search spaces (Platt et al., 2006), and mobile robot navigation (Dorigo and Colombetti, 1993).”
Learning Multi-Arm Manipulation Through Collaborative Teleoperation (2020, background): “Multi-Agent Reinforcement Learning: Multi-Agent Re-inforcement Learning [5, 35] in cooperative settings has been widely studied [11, 12, 14, 21, 26, 34], and applied to domains such as video games [28] and visual question answering [9].”
Learning Multiagent Communication with Backpropagation (2016, background): “games [31], but they assume full visibility of the environment and lack communication. There is a rich literature on multi-agent reinforcement learning (MARL) [2], particularly in the robotics domain [19, 27, 6, 22, 3]. Amongst fully cooperative algorithms, many approaches [13, 16, 35] avoid the nee”
Multi-agent reinforcement learning as a rehearsal for decentralized planning (2016, neutral): “This work also differs from the large body of work on reward shaping where an agent explicitly engineers the reward function [17,18] to improve learning speed, or a human feedback is used to construct shaping rewards [13].”
Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks (2016, background): “Such problems arise naturally in a variety of settings, such as multi-robot systems and sensor networks (Matari, 1997; Fox et al., 2000; Gerkey & Matari, 2004; Olfati-Saber et al., 2007; Cao et al., 2013).”
Hierarchical multi-agent reinforcement learning (2001, background): “In addition, since agents carry out actions in parallel, the environment is usually non-stationary and often non-Markovian as well [9].”
Cooperative Multi-Agent Learning: The State of the Art (2005, methodology): “Balch [9] uses a shaped reinforcement reward function (earlier suggested by Mataric [161]) which depends on the number of partial steps fulfilled towards accomplishing the joint task.”
Multiagent Reinforcement Learning for Multi-Robot Systems: A Survey (2004, methodology): “The work in [2] presented a formulation of RL that enables learning in the concurrent multi-robot domain.”
Teachable robots: Understanding human teaching behavior to build more effective robot learners (2008, background): “Various works have addressed some of the hard problems robots face when learning in the real-world, e.g., real-time learning in environments that are partially observable, dynamic, continuous (Mataric, 1997; Thrun and Mitchell, 1993; Thrun, 2002).”
Training an Actor-Critic Reinforcement Learning Controller for Arm Movement Using Human-Generated Rewards (2017, background): “While decreasing the frequency of rewards has been shown to slow learning speed [21], [43], techniques such as increasing the actor’s learning rate [21] have been found to compensate for the slowed learning that results from sparse rewards.”
Improved Learning of Dynamics Models for Control (2016, background): “Learning based approaches for controlling robotic systems, or more generally autonomous agents, fall into two primary categories: model-based [1–3] and model-free [4–7].”
Deep-Learned Collision Avoidance Policy for Distributed Multiagent Navigation (2016, methodology): “In this paper, we use an end-to-end framework to learn a collision-avoidance policy which is robust to the imperfect sensing and requires no inter-agent communication, and thus is still fully distributed.”
Efficient Distributed Reinforcement Learning through Agreement (2008, background): “, [9, 4]), techniques that rely on the Markov property of the world.”
Analysis of Dynamic Task Allocation in Multi-Robot Systems (2006, background): “The few application-level studies to consider adaptation and learning in multi-robot systems [13, 26, 23, 12, 4].”
Theoretical considerations of potential-based reward shaping for multi-agent systems (2011, methodology): “One such method, empirically demonstrated to decrease the time for each individual learning in a common environment to converge on a stable policy, is incorporating heuristic knowledge [17, 25].”