Robustness analysis of evolutionary controller tuning using real systems

Góngora, Mario; Passow, Benjamin N.; Hopgood, Adrian A.

doi:10.1109/cec.2009.4983001

Cited by 12 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A gap can even exist when the controllers are directly evolved on the real system, if the experimental set-up which allows to evaluate the individuals is too different from the real environment of the robot. It has notably been observed in [19] on a small helicopter.…”

Section: Introductionmentioning

confidence: 94%

“…Consequently, the few works in which controllers have been directly evolved on the robot often optimized few individuals during few generations, which reduces the efficiency of the evolutionary methods. For instance in [19], controllers for a small helicopter have been evolved with a population of 20 individuals during 30 generations, with few minutes between generations to avoid over-heating, that is only 600 evaluations during the optimization process. In [52], optimization has directly been applied to a prototype ornithopter machine to maximize its lift with 3000 evaluations on the physical system during the optimization, which seems more consistent with Sylvain Koos, Jean-Baptiste Mouret and Stéphane Doncieux are with the ISIR, CNRS UMR 7222, Université Pierre et Marie Curie, F-75005, Paris, France.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics

Koos

Mouret

Doncieux

2013

IEEE Trans. Evol. Computat.

215

184

View full text Add to dashboard Cite

Abstract-The reality gap, that often makes controllers evolved in simulation inefficient once transferred onto the physical robot, remains a critical issue in Evolutionary Robotics (ER). We hypothesize that this gap highlights a conflict between the efficiency of the solutions in simulation and their transferability from simulation to reality: the most efficient solutions in simulation often exploit badly modeled phenomena to achieve high fitness values with unrealistic behaviors. This hypothesis leads to the Transferability approach, a multi-objective formulation of ER in which two main objectives are optimized via a Pareto-based Multi-Objective Evolutionary Algorithm: (1) the fitness and (2) the transferability, estimated by a simulation-to-reality (STR) disparity measure. To evaluate this second objective, a surrogate model of the exact STR disparity is built during the optimization. This Transferability approach has been compared to two realitybased optimization methods, a noise-based approach inspired from Jakobi's minimal simulation methodology and a local search approach. It has been validated on two robotic applications: 1) a navigation task with an e-puck robot; 2) a walking task with an 8-DOF quadrupedal robot. For both experimental set-ups, our approach successfully finds efficient and well-transferable controllers only with about ten experiments on the physical robot.

show abstract

Section: Introductionmentioning

confidence: 94%

Section: Introductionmentioning

confidence: 99%

The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics

Koos

Mouret

Doncieux

2013

IEEE Trans. Evol. Computat.

215

184

View full text Add to dashboard Cite

show abstract

“…Evolving in hardware bypasses the problem of the reality gap completely, and if evolution is performed on the unrestricted system in the environment where it will be serving, also bypasses the problem sometimes seen in simplified or limited experiments in hardware as well [9]. Evolution in hardware is most often done off-line to perform a one-time adaptation to a new task or environment, but can also been done constantly in an on-line fashion to continuously adapt to both slow and abrupt changes to the robot itself or its environment [6].…”

Section: Real World Evolutionmentioning

confidence: 99%

Real-world evolution adapts robot morphology and control to hardware limitations

Nygaard

Martín

Samuelsen

et al. 2018

Proceedings of the Genetic and Evolutionary Computation Conference

View full text Add to dashboard Cite

For robots to handle the numerous factors that can affect them in the real world, they must adapt to changes and unexpected events. Evolutionary robotics tries to solve some of these issues by automatically optimizing a robot for a specific environment. Most of the research in this field, however, uses simplified representations of the robotic system in software simulations. The large gap between performance in simulation and the real world makes it challenging to transfer the resulting robots to the real world. In this paper, we apply real world multi-objective evolutionary optimization to optimize both control and morphology of a four-legged mammal-inspired robot. We change the supply voltage of the system, reducing the available torque and speed of all joints, and study how this affects both the fitness, as well as the morphology and control of the solutions. In addition to demonstrating that this realworld evolutionary scheme for morphology and control is indeed feasible with relatively few evaluations, we show that evolution under the different hardware limitations results in comparable performance for low and moderate speeds, and that the search achieves this by adapting both the control and the morphology of the robot.

show abstract

“…A blimp controller is successfully evolved [12], but the slow dynamics of the blimp simpli es recovery from dangerous states. Control of a miniature helicopter [14] is evolved, although only height and yaw control are optimized.…”

Section: Er With Flying Robotsmentioning

confidence: 99%

On Self-Adaptive Mutation Restarts for Evolutionary Robotics with Real Rotorcraft

Howard

2017

Preprint

View full text Add to dashboard Cite

Self-adaptive parameters are increasingly used in the eld of Evolutionary Robotics, as they allow key evolutionary rates to vary autonomously in a context-sensitive manner throughout the optimisation process. A signi cant limitation to self-adaptive mutation is that rates can be set unfavourably, which hinders convergence. Rate restarts are typically employed to remedy this, but thus far have only been applied in Evolutionary Robotics for mutation-only algorithms. is paper focuses on the level at which evolutionary rate restarts are applied in population-based algorithms with >1 evolutionary operator. A er testing on a real hexacopter hovering task, we conclude that individual-level restarting results in higher tness solutions without tness stagnation, and population restarts provide a more stable rate evolution. Without restarts, experiments can become stuck in suboptimal controller/rate combinations which can be di cult to escape from.

show abstract

Robustness analysis of evolutionary controller tuning using real systems

Cited by 12 publications

References 20 publications

The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics

The Transferability Approach: Crossing the Reality Gap in Evolutionary Robotics

Real-world evolution adapts robot morphology and control to hardware limitations

On Self-Adaptive Mutation Restarts for Evolutionary Robotics with Real Rotorcraft

Contact Info

Product

Resources

About