Crossing the reality gap in evolutionary robotics by promoting transferable controllers

By Sylvain Koos, Jean-Baptiste Mouret, Stéphane Doncieux. 2010

138 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biomedical 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 Biomedical Engineering

Citations in the corpus

Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience (2018, methodology): “resentations of real and synthetic data [16], using generative models to make synthetic images more realistic [17], fine-tuning the policies trained in simulation in the real world [18], learning inverse dynamics models [19], multiobjective optimization of task fitness and transferability to the rea”
Sim-to-Real: Learning Agile Locomotion For Quadruped Robots (2018, background): “This reality gap [23, 3] is caused by model discrepancies between the simulated and the real physical system.”
Preparing for the Unknown: Learning a Universal Policy with Online System Identification (2017, background): “The discrepancy between what can be achieved in simulation and that in real world is referred as the “Reality Gap” in the Evolutionary Robotics community [22, 14].”
Off-Policy Evaluation via Off-Policy Classification (2019, methodology): “While to our knowledge no prior work has proposed a classiﬁcation-based OPE approach, several prior works have used supervised classiﬁers to predict transfer performance from a few runs in the test environment [17, 18].”
RMA: Rapid Motor Adaptation for Legged Robots (2021, background): “However, such policies are difﬁcult to transfer to the real world [31, 39, 6].”
PAC Reinforcement Learning With an Imperfect Model (2018, background): “A promising approach to addressing this issue is to train an RL agent in a simulator and transfer the solution to the real environment, which is particularly relevant but not limited to robotics domains (Koos, Mouret, and Doncieux 2010; Cutler and How 2015; Hanna and Stone 2017).”
Sim2Real Transfer for Deep Reinforcement Learning with Stochastic State Transition Delays (2020, background): “The resulting modeling discrepancies cause a reality gap , which makes the transfer of RL policies from simulation to the real-world (Sim2Real) a challenge [4].”
Humanoid robots learning to walk faster: from the real world to simulation and back (2013, background): “[16] made a recent effort to answer this question by studying a quantity they labeled “Transferability”, which is a measure on how well each component of a simulated model’s performance transfers from the simulation to reality.”
The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics (2013, methodology): “For this second application, preliminary results have been published in [35].”
Interactive reinforcement learning for task-oriented dialogue management (2016, methodology): “These approaches, although intricate, are lossy models of human conversational behavior that lead to a "reality gap" between simulation and real-world usage of the dialogue manager, which similarly afflicts the domain of robot control (Koos et al. (2010)).”