Deep reinforcement learning for high precision assembly tasks

Inoue, Takuya; Magistris, Giovanni De; Munawar, Asim; Yokoya, Tsuyoshi; Tachibana, Ryuki

doi:10.1109/iros.2017.8202244

Cited by 284 publications

(171 citation statements)

References 14 publications

Supporting

Mentioning

169

Contrasting

Unclassified

Order By: Relevance

“…A recently developed model-based reinforcement learning algorithm called guided policy search (GPS) provided new insights into training end-to-end policy for solving contactrich manipulation problems [15,26,27,4,28]; however; this method is not suitable for this high-precision setting because it has no means of avoiding local optima by its formulation. There are also approaches tackling this problem by explicitly modeling contact dynamics [29,30,31,32,33] Inoue et al [34] use LSTM to learn two separate policies for finding and inserting a peg into a hole; however, their methods require several pre-defined heuristics, and also the action space is discrete.…”

Section: Problem Statement and Related Workmentioning

confidence: 99%

Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assembly

Luo

Solowjow

Wen

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

155

View full text Add to dashboard Cite

Precise robotic manipulation skills are desirable in many industrial settings, reinforcement learning (RL) methods hold the promise of acquiring these skills autonomously. In this paper, we explicitly consider incorporating operational space force/torque information into reinforcement learning; this is motivated by humans heuristically mapping perceived forces to control actions, which results in completing high-precision tasks in a fairly easy manner. Our approach combines RL with force/torque information by incorporating a proper operational space force controller; where we also exploit different ablations on processing this information. Moreover, we propose a neural network architecture that generalizes to reasonable variations of the environment. We evaluate our method on the open-source Siemens Robot Learning Challenge, which requires precise and delicate force-controlled behavior to assemble a tight-fit gear wheel set. a) Robot learning for complex assemblies. b) Assembled gear.

show abstract

Section: Problem Statement and Related Workmentioning

confidence: 99%

Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assembly

Luo

Solowjow

Wen

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

155

View full text Add to dashboard Cite

show abstract

“…Reinforcement learning has been applied successfully in simulated and real-world robotic manipulation [3], [4], [5], [8], [9], [10], locomotion [6], [2] and autonomous vehicles [11]. Many of the demonstrated scenarios used tailored policy representations or discretized action spaces.…”

Section: Related Workmentioning

confidence: 99%

Learning Robust Manipulation Skills with Guided Policy Search via Generative Motor Reflexes

Ennen

Bresenitz

Vossen

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Guided Policy Search enables robots to learn control policies for complex manipulation tasks efficiently. Therein, the control policies are represented as high-dimensional neural networks which derive robot actions based on states. However, due to the small number of real-world trajectory samples in Guided Policy Search, the resulting neural networks are only robust in the neighbourhood of the trajectory distribution explored by real-world interactions. In this paper, we present a new policy representation called Generative Motor Reflexes, which is able to generate robust actions over a broader state space compared to previous methods. In contrast to prior state-action policies, Generative Motor Reflexes map states to parameters for a state-dependent motor reflex, which is then used to derive actions. Robustness is achieved by generating similar motor reflexes for many states. We evaluate the presented method in simulated and real-world manipulation tasks, including contact-rich peg-in-hole tasks. Using these evaluation tasks, we show that policies represented as Generative Motor Reflexes lead to robust manipulation skills also outside the explored trajectory distribution with less training needs compared to previous methods.1 Philipp Ennen, Pia Bresenitz, Rene Vossen, Frank Hees are with the

show abstract

“…This is particularly true in the case of data-center optimization since heat distribution occurs over time (e.g., a given command may not impact sensor readings instantaneously) and can depend on non-controllable parameters (e.g., weather conditions). DRL approaches have recently provided considerable results on such problems, where the relationship between states and optimal actions is difficult to model formally, e.g., playing Atari video games directly from pixels using convolutional neural networks (CNNs) rather than handcrafted features [7], or learning complex robotic tasks in both simulated [3] and real environments [5]. In DRL, control policies π θ are typically represented by deep neural networks parameterized by vector of parameters θ (e.g., neuron weights and biases).…”

Section: Model-free Reinforcement Learningmentioning

confidence: 99%

Reinforcement Learning Testbed for Power-Consumption Optimization

Moriyama

Magistris

Tatsubori

et al. 2018

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

Common approaches to control a data-center cooling system rely on approximated system/environment models that are built upon the knowledge of mechanical cooling and electrical and thermal management. These models are difficult to design and often lead to suboptimal or unstable performance. In this paper, we show how deep reinforcement learning techniques can be used to control the cooling system of a simulated data center. In contrast to common control algorithms, those based on reinforcement learning techniques can optimize a system's performance automatically without the need of explicit model knowledge. Instead, only a reward signal needs to be designed. We evaluated the proposed algorithm on the open source simulation platform EnergyPlus. The experimental results indicate that we can achieve 22% improvement compared to a model-based control algorithm built into the EnergyPlus. To encourage the reproduction of our work as well as future research, we have also publicly released an open-source EnergyPlus wrapper interface 1 directly compatible with existing reinforcement learning frameworks.

show abstract

Deep reinforcement learning for high precision assembly tasks

Cited by 284 publications

References 14 publications

Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assembly

Reinforcement Learning on Variable Impedance Controller for High-Precision Robotic Assembly

Learning Robust Manipulation Skills with Guided Policy Search via Generative Motor Reflexes

Reinforcement Learning Testbed for Power-Consumption Optimization

Contact Info

Product

Resources

About