Learning to Control in Operational Space

By Jan Peters, Stefan Schaal. 2008

196 citations recorded by the source corpus.

Subfields: 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 Control and Systems Engineering

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, methodology): “…of additional structure within the reinforcement learning domain, including, for instance, the (approximate) Markov structure of problems, developing approximate models, value-function estimates when available (Peters and Schaal, 2008c), or even simply the causal ordering of actions and rewards.”
On-line regression algorithms for learning mechanical models of robots: A survey (2011, methodology): “The reader can refer to Schaal et al. [90] or Vijayakumar & Schaal [108] for a detailed presentation of the incremental version of the algorithm.”
Reinforcement learning for robot soccer (2009, methodology): “…including autonomously flying artistic maneuvers with a helicopter (Bagnell and Schneider 2001; Ng et al. 2004), teaching a robot to hit a baseball (Peters and Schaal 2006), operational space control (Peters and Schaal 2008a) and learning ball-in-a-cup with a robot arm (Kober et al. 2008).”
Reinforcement learning of motor skills in high dimensions: A path integral approach (2010, methodology): “…from statistical learning and empirical inference with classical RL approaches in order to minimize tuning parameters and numerical problems, such that ultimately more efﬁcient algorithms can be developed that scale to signiﬁcantly more complex learning systems [3], [4], [5], [6], [7], [8], [9].”
Learning Forward Models for the Operational Space Control of Redundant Robots (2010, methodology): “However, the learning of this mapping is out of the scope of the work presented here (interested readers can refer to Peters and Schaal (2008) and Nguyen-Tuong et al (2008)) and it is supposed”
Imitation and Reinforcement Learning (2010, methodology): “However, RL for MPs is a very specific type of learning problem, where it is hard to apply generic algorithms [4], [15] because of the high-dimensional states s 1⁄4 1⁄2z, x that defy discretization of the state space.”
A Generalized Path Integral Control Approach to Reinforcement Learning (2010, unknown): “…1 The optimal control solution to the stochastic optimal control problem expressed by (1),(2),(3) and (4) is formulated as: uti = lim dt→0 [ −R−1G(c)ti T ∫ p̃ (τ i)∇x(c)ti S̃(τ i)dτ i ] , where p̃ (τ i) = exp (− 1λ S̃(τ i))∫ exp (− 1λ S̃(τ i))dτ i is a path dependent probability distribution.”
Model learning for robot control: a survey (2011, background): “In (Peters and Schaal 2008; Ting et al. 2008), the authors attempt to directly approximate the multi-valued relationship using local learning approaches.”
Learning task constraints in operational space formulation (2017, background): “This includes learning the operational space control law [9], or recovering the underlying redundancy from the constrained data [10], [11].”
Deep auto-encoder neural networks in reinforcement learning (2010, background): “Recently, there have been reported several impressive successes of applying reinforcement learning to real-world systems [1], [2], [3].”
Task-Specific Generalization of Discrete and Periodic Dynamic Movement Primitives (2010, background): “2These problems can be addressed by biasing [50] or re-weighting [51] the data.”
Gaussian process dynamic programming (2009, methodology): “Traditional policy learning meth-ods as discussed by Peters and Schaal (2008a,b,c) or standard GP models with smoothness-favoring covariance functions, which have been used for instance by Ras-mussen and Deisenroth (2008), are inappropriate to model discontinuities.”
Learning Policy Improvements with Path Integrals (2010, background): “It also makes an interesting connection to previous work on RL based on probability matching [7, 8, 2] and explains why probability matching algorithms can be successful”
Variational Bayesian learning of nonlinear hidden state-space models for model predictive control (2009, background): “Application of [30] to robotics is given in [24].”
Curiosity driven reinforcement learning for motion planning on humanoids (2014, methodology): “, 2001), and operational space control laws (Peters and Schaal, 2008), but these are not curiosity-driven.”