Adapting the control systems of robots on the fly is important in robotic systems of the future. In this paper we present and investigate a three-fold adaptive system based on evolution, individual and social learning...
详细信息
ISBN:
(纸本)9781450334723
Adapting the control systems of robots on the fly is important in robotic systems of the future. In this paper we present and investigate a three-fold adaptive system based on evolution, individual and social learning in a group of robots and report on a proof-of-concept study based on e-pucks. We distinguish inheritable and learnable components in the robots' makeup, specify and implement operators for evolution, learning and social learning, and test the system in an arena where the task is to learn to avoid obstacles. In particular, we make the sensory layout evolvable, the locomotion control system learnable and investigate the effects of including social learning in the 'adaptation engine'. Our simulation experiments demonstrate that the full mix of three adaptive mechanisms is practicable and that adding social learning leads to better controllers faster.
Traditional evolutionary algorithms tend to converge to a single good solution, which can limit their chance of discovering more diverse and creative outcomes. Divergent search, on the other hand, aims to counter conv...
详细信息
ISBN:
(纸本)9781450356183
Traditional evolutionary algorithms tend to converge to a single good solution, which can limit their chance of discovering more diverse and creative outcomes. Divergent search, on the other hand, aims to counter convergence to local optima by avoiding selection pressure towards the objective. Forms of divergent search such as novelty or surprise search have proven to be beneficial for both the efficiency and the variety of the solutions obtained in deceptive tasks. Importantly for this paper, early results in maze navigation have shown that combining novelty and surprise search yields an even more effective search strategy due to their orthogonal nature. Motivated by the largely unexplored potential of coupling novelty and surprise as a search strategy, in this paper we investigate how fusing the two can affect the evolution of soft robot morphologies. We test the capacity of the combined search strategy against objective, novelty, and surprise search, by comparing their efficiency and robustness, and the variety of robots they evolve. Our key results demonstrate that novelty-surprise search is generally more efficient and robust across eight different resolutions. Further, surprise search explores the space of robot morphologies more broadly than any other algorithm examined.
In this paper, we explore how robots in a swarm can individually exploit collisions to produce self-organizing behaviours at the macroscopic scale. We propose to focus on two behaviours that modify the orientation of ...
详细信息
ISBN:
(纸本)9781450392372
In this paper, we explore how robots in a swarm can individually exploit collisions to produce self-organizing behaviours at the macroscopic scale. We propose to focus on two behaviours that modify the orientation of a robot during a collision, which are inspired by positive and negative feedback observed in Nature. These two behaviours differ in the nature of the feedback produced after a collision by favouring either (1) the alignment or (2) the anti-alignment of the robot with an external force, whether it is an obstacle or another robot. We describe a social learning algorithm using evolutionary operators to learn individual policies that exploit these behaviours in an online and distributed fashion. This algorithm is validated both in simulation and with real robots to solve two tasks involving phototaxis, one of which requires self-organized aggregation to be completed.
Cyclic genetic algorithms were developed to evolve single loop control programs for robots. These programs have been used for three levels of control: individual leg movement, gait generation, and area search path fin...
详细信息
ISBN:
(纸本)078039044X
Cyclic genetic algorithms were developed to evolve single loop control programs for robots. These programs have been used for three levels of control: individual leg movement, gait generation, and area search path finding. In all of these applications the cyclic genetic algorithm learned the cycle of actuator activations that could be continually repeated to produce the desired behavior. Although very successful for these applications, it was not applicable to control problems that required different behaviors in response to sensor inputs. Control programs for this type of behavior require multiple loops with conditional statements to regulate the branching. In this paper, we present modifications to the standard cyclic genetic algorithm that allow it to learn multi-loop control programs that can react to sensor input.
Bootstrapped Neuro-Simulation is a technique in which a Neural Network Simulator, used for the evaluation of controllers during the evolutionary robotics process, is trained concurrently with the evolution of the cont...
详细信息
ISBN:
(数字)9781665467087
ISBN:
(纸本)9781665467087
Bootstrapped Neuro-Simulation is a technique in which a Neural Network Simulator, used for the evaluation of controllers during the evolutionary robotics process, is trained concurrently with the evolution of the controllers themselves. This removes the need for creating a simulator before commencing controller evolution and results in a simulator that is not only tailored specifically to the robot being used, but also to the task that the robot is expected to perform. This paper demonstrates that Bootstrapped Neuro-Simulation can also be used for damage recovery since the Neural Network Simulator adapts to physical changes to the robot and enables the evolution of controllers that utilize the undamaged components of the robot. Limbs of a hexapod robot are disabled to simulate damage in the experiments described in this paper. Various adaptations to the Bootstrapped Neuro-Simulation algorithm are investigated in simulation. A real-world robot is used to demonstrate the successful recovery from damage and to illustrate situations where the adaptations were found to be beneficial.
In order to evolve large robot controllers for increasingly complex tasks, fully connected neural networks are not feasible. However, manually designing sparse neural connectivity is not intuitive, and thus should be ...
详细信息
ISBN:
(纸本)9781450305570
In order to evolve large robot controllers for increasingly complex tasks, fully connected neural networks are not feasible. However, manually designing sparse neural connectivity is not intuitive, and thus should be placed under evolutionary control. Here I show how spontaneous structural modularity can arise in the connectivity of evolved robot controllers if the controllers are boolean networks, and are selected to converge on point attractors that correspond to successful robot behaviors.
In evolutionary robotics, evolutionary methods are used to optimize robots to different tasks. Because using physical robots is costly in terms of both time and money, simulated robots are generally used instead. Most...
详细信息
ISBN:
(纸本)9781450349390
In evolutionary robotics, evolutionary methods are used to optimize robots to different tasks. Because using physical robots is costly in terms of both time and money, simulated robots are generally used instead. Most physics engines are written in C++ which can be a barrier for new programmers. In this paper we present two Python wrappers, Pyrosim and Evosoro, around two well used simulators, Open Dynamics Engine (ODE) and Voxelyze/VoxCAD, which respectively handle rigid and soft bodied simulation. Python is an easier language to understand so more time can be spent on developing the actual experiment instead of programming the simulator.
Novelty Search, a new type of evolutionary Algorithm, has shown much promise in the last few years. Instead of selecting for phenotypes that are closer to an objective, Novelty Search assigns rewards based on how diff...
详细信息
ISBN:
(纸本)9781450326629
Novelty Search, a new type of evolutionary Algorithm, has shown much promise in the last few years. Instead of selecting for phenotypes that are closer to an objective, Novelty Search assigns rewards based on how different the phenotypes are from those already generated. A common criticism of Novelty Search is that it is effectively random or exhaustive search because it tries solutions in an unordered manner until a correct one is found. Its creators respond that over time Novelty Search accumulates information about the environment in the form of skills relevant to reaching uncharted territory, but to date no evidence for that hypothesis has been presented. In this paper we test that hypothesis by transferring robots evolved under Novelty Search to new environments (here, mazes) to see if the skills they've acquired generalize. Three lines of evidence support the claim that Novelty Search agents do indeed learn general exploration skills. First, robot controllers evolved via Novelty Search in one maze and then transferred to a new maze explore significantly more of the new environment than nonevolved (randomly generated) agents. Second, a Novelty Search process to solve the new mazes works significantly faster when seeded with the transferred controllers versus randomly-generated ones. Third, no significant difference exists when comparing two types of transferred agents: those evolved in the original maze under (1) Novelty Search vs. (2) a traditional, objective-based fitness function. The evidence gathered suggests that, like traditional evolutionary Algorithms with objective-based fitness functions, Novelty Search is not a random or exhaustive search process, but instead is accumulating information about the environment, resulting in phenotypes possessing skills needed to explore their world.
This paper presents a study of the efficacy of comparative controller design methods that aim to produce generalised problem solving behaviours. In this case study, the goal was to use neuro-evolution to evolve genera...
详细信息
ISBN:
(纸本)9783319165486;9783319165493
This paper presents a study of the efficacy of comparative controller design methods that aim to produce generalised problem solving behaviours. In this case study, the goal was to use neuro-evolution to evolve generalised maze solving behaviours. That is, evolved robot controllers that solve a broad range of mazes. To address this goal, this study compares objective, non-objective and hybrid approaches to direct the search of a neuro-evolution controller design method. The objective based approach was a fitness function, the non-objective based approach was novelty search, and the hybrid approach was a combination of both. Results indicate that, compared to the fitness function, the hybrid and novelty search evolve significantly more maze solving behaviours that generalise to larger and more difficult maze sets. Thus this research provides empirical evidence supporting novelty and hybrid novelty-objective search as approaches for potentially evolving generalised problem solvers.
Exploration and exploitation are two complementary aspects of evolutionary Algorithms. Exploration, in particular, is promoted by specific diversity keeping mechanisms generally relying on the genotype or on the fitne...
详细信息
ISBN:
(纸本)9781450305570
Exploration and exploitation are two complementary aspects of evolutionary Algorithms. Exploration, in particular, is promoted by specific diversity keeping mechanisms generally relying on the genotype or on the fitness value. Recent works suggest that, in the case of evolutionary robotics or more generally behavioral system evolution, promoting exploration directly in the behavioral space is of critical importance. In this work an exploration indicator is proposed, based on the sparseness of the population in the behavioral space. This exploration measure is used on two challenging neuro-evolution experiments and validated by showing the dependence of the fitness at the end of the run on the exploration measure during the very first generations. Such a prediction ability could be used to design parameter settings algorithms or selection algorithms dedicated to the evolution of behavioral systems. Several other potential uses of this measure are also proposed and discussed.
暂无评论