Noise and the Reality Gap: The Use of Simulation in Evolutionary Robotics

By N. Jakobi, P. Husbands, Inman R. Harvey. 1995

How researchers see this paper

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

Citations in the corpus

Reinforcement learning in robotics: A survey (2013, background): “adding a little noise will smooth model errors and avoid policy over-fitting (Jakobi et al., 1995; Atkeson, 1998).”
Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience (2018, background): “learned in simulations often cannot be directly applied on real world systems, a phenomenon also known as the reality gap [5].”
Evolutionary design and behavior analysis of neuromodulatory neural networks for mobile robots control (2007, background): “To tackle the problem, many researchers have proposed to use techniques such as fitness-averaging [3] and/or adding noise [4,5] in the evolutionary process.”
Learning reactive and planning rules in a motivationally autonomous animat (1996, result): “TheMIT Press/Bradford Books, 1986[26] N. Jakobi, P. Husbands and I. Harvey, "Noise and the RealityGap: The Use of Simulation in Evolutionary Robotics," in Pro-ceedings of the Third European Conference on Arti cial Life,in Press, 1995.”
Reinforcement Learning for UAV Attitude Control (2018, background): “Known as the reality gap, transferring from simulation to the real-world has been researched extensively as being problematic without taking additional steps to increase realism in the simulator [19, 30].”
An Introduction to Deep Reinforcement Learning (2018, unknown): “When first learning in a simulation, the difference with the real-world domain is known as the reality gap (see e.g. Jakobi et al., 1995).”
Leveraging multiple simulators for crossing the reality gap (2012, methodology): “(2) Where, D is the drag force ρ is the liquid density A is the frontal area Cd is the drag coefficient Cl is the lift coefficient and V is the relative velocity The buoyancy force can be calculated from Archimedes principle.”
Swarm Robotics: Past, Present, and Future (2021, background): “The problem, known in robotics as the simulation–reality gap [62], is that differences between the models used in simulation and their real-world counterparts cause a drop in performance when robot”
SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining (2021, unknown): “However, although generated images may look remarkably realistic, they still su ﬀ er from a “reality gap” [41].”
A Systematic Review of Perception System and Simulators for Autonomous Vehicles Research (2019, background): “However, for a simulation to be the most identical to reality, perfect models are needed to minimize the reality gap [90,91].”
On the use of simulation in robotics: Opportunities, challenges, and suggestions for moving forward (2020, background): “For instance, in machine learning, sophisticated models are expected to better negotiate the simulation-to-reality gap, i.e., to yield simulation-learned control policies that work well on the real robot (4).”
DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics (2020, methodology): “are usually used in simulation, which is possible in the present context considering the knowledge acquired from the bootstrap phase, but introduces a further challenge in the form of the reality gap [76], where policies learnt in simulation must be transferred to the real robot. Our approach [73] u”