Search

The use of evolutionary robotic systems to model aspects of evolutionary biology is well-established. Yet, few studies have asked the question, "What kind of model is an evolutionary robotic system?" This paper seeks to address that question in several ways. First, it is addressed by applying a structured model description developed for physical robot models of animal sensorimotor systems, then by outlining the strengths and limitations of evolutionary robotics for modelling evolutionary biology, and, finally, by considering the deepest questions in evolution and which of them might feasibly be modelled by evolutionary robotics. The paper concludes that although evolutionary robotics faces serious limitations in exploring deeper questions in evolutionary biology, its bottom-up approach to modelling populations of evolving phenotypes and their embodied interactions holds significant value for both testing and generating hypotheses. [ABSTRACT FROM AUTHOR]

The discrepancy between simulated and hardware experiments, the reality gap, is a challenge in evolutionary robotics. While strategies have been proposed to address this gap in fixed-body robots, they are not viable when dealing with populations and generations where the body is in constant change. The continual evolution of body designs necessitates the manufacturing of new robotic structures, a process that can be time-consuming if carried out manually. Moreover, the increased manufacturing time not only prolongs hardware experimental durations but also disrupts the synergy between hardware and simulated experiments. Failure to effectively manage these challenges could impede the implementation of evolutionary robotics in real-life environments. The Autonomous Robot Evolution project presents a framework to tackle these challenges through a case study. This paper describes the main three contributions of this work: Firstly, it analyses the different reality gap experienced by each different robot or the heterogenous reality gap. Secondly, it emphasizes the importance of automation in robot manufacturing. And thirdly, it highlights the necessity of a framework to orchestrate the synergy between simulated and hardware experiments. In the long term, integrating these contributions into evolutionary robotics is envisioned to enable the continuous production of robots in real-world environments. [ABSTRACT FROM AUTHOR]

Modular robots are collections of simple embodied agents, the modules, that interact with each other to achieve complex behaviors. Each module may have a limited capability of perceiving the environment and performing actions; nevertheless, by behaving coordinately, and possibly by sharing information, modules can collectively perform complex actions. In principle, the greater the actuation, perception, and communication abilities of the single module are the more effective is the collection of modules. However, improved abilities also correspond to more complex controllers and, hence, larger search spaces when designing them by means of optimization. In this article, we analyze the impact of perception, actuation, and communication abilities on the possibility of obtaining good controllers for simulated modular robots, that is, controllers that allow the robots to exhibit collective intelligence. We consider the case of modular soft robots, where modules can contract, expand, attach, and detach from each other, and make them face two tasks (locomotion and piling), optimizing their controllers with evolutionary computation. We observe that limited abilities often do not prevent the robots from succeeding in the task, a finding that we explain with (a) the smaller search space corresponding to limited actuation, perception, and communication abilities, which makes the optimization easier, and (b) the fact that, for this kind of robot, morphological computation plays a significant role. Moreover, we discover that what matters more is the degree of collectivity the robots are required to exhibit when facing the task. [ABSTRACT FROM AUTHOR]

The use of evolutionary robotic systems to model aspects of evolutionary biology is well-established. Yet, few studies have asked the question, “What kind of model is an evolutionary robotic system?” This paper seeks to address that question in several ways. First, it is addressed by applying a structured model description developed for physical robot models of animal sensorimotor systems, then by outlining the strengths and limitations of evolutionary robotics for modelling evolutionary biology, and, finally, by considering the deepest questions in evolution and which of them might feasibly be modelled by evolutionary robotics. The paper concludes that although evolutionary robotics faces serious limitations in exploring deeper questions in evolutionary biology, its bottom-up approach to modelling populations of evolving phenotypes and their embodied interactions holds significant value for both testing and generating hypotheses.

Exposing an evolutionary algorithm that is used to evolve robot controllers to variable conditions is necessary to obtain solutions which are robust and can cross the reality gap. However, we do not yet have methods for analyzing and understanding the impact of the varying morphological conditions which impact the evolutionary process, and therefore for choosing suitable variation ranges. By morphological conditions, we refer to the starting state of the robot, and to variations in its sensor readings during operation due to noise. In this paper, we introduce a method that permits us to measure the impact of these morphological variations and we analyze the relation between the amplitude of variations, the modality with which they are introduced, and the performance and robustness of evolving agents. Our results demonstrate that (i) the evolutionary algorithm can tolerate morphological variations which have a very high impact, (ii) variations affecting the actions of the agent are tolerated much better than variations affecting the initial state of the agent or of the environment, and (iii) improving the accuracy of the fitness measure through multiple evaluations is not always useful. Moreover, our results show that morphological variations permit generating solutions which perform better both in varying and non-varying conditions. [ABSTRACT FROM AUTHOR]

In this paper, we compare Bayesian Optimization, Differential Evolution, and an Evolution Strategy employed as a gait-learning algorithm in modular robots. The motivational scenario is the joint evolution of morphologies and controllers, where "newborn" robots also undergo a learning process to optimize their inherited controllers (without changing their bodies). This context raises the question: How do gait-learning algorithms compare when applied to various morphologies that are not known in advance (and thus need to be treated as without priors)? To answer this question, we use a test suite of twenty different robot morphologies to evaluate our gait-learners and compare their efficiency, efficacy, and sensitivity to morphological differences. The results indicate that Bayesian Optimization and Differential Evolution deliver the same solution quality (walking speed for the robot) with fewer evaluations than the Evolution Strategy. Furthermore, the Evolution Strategy is more sensitive for morphological differences (its efficacy varies more between different morphologies) and is more subject to luck (repeated runs on the same morphology show greater variance in the outcomes). [ABSTRACT FROM AUTHOR]

In this paper, different test cases where robotic arms are tested will be presented. A robotic arm is tested for the gravity effects that can be observed on it. The other robotic arm is tested for how much precision it has by using it to learn to write. The other robotic arm is tested on how well it can function as a solar tracker and how precisely it can function as an energy harvester. On the basis of these tests, the robotic arm's mechanical structure, electronics, and software are put to the test. The software is based on evolutionary software that implements genetic algorithms. The entire command system is also ported to FPGAs (to hardware) to increase speed and response time. [ABSTRACT FROM AUTHOR]

In the field of evolutionary robotics, choosing the correct genetic representation is a complicated and delicate matter, especially when robots evolve behaviour and morphology at the same time. One principal problem is the lack of methods or tools to investigate and compare representations. In this paper we introduce and evaluate such a tool based on the biological notion of heritability. Heritability captures the proportion of phenotypic variation caused by genotypic variation and is often used to better understand the transmissibility of traits in real biological systems. As a proof of concept, we compare the heritability of various robot traits in two systems, one using a direct (tree based) representation and one using an indirect (grammar based) representation. We measure changes in heritability during the course of evolution and investigate how direct and indirect representation can be biased towards more exploration or exploitation throughout the course of evolution. The empirical study shows that heritability can be a useful tool to analyze different representations without running complete evolutionary processes using them. [ABSTRACT FROM AUTHOR]

Biological agents have bodies that are composed mostly of soft tissue. Researchers have resorted to soft bodies to investigate Artificial Life (ALife)-related questions; similarly, a new era of soft-bodied robots has just begun. Nevertheless, because of their infinite degrees of freedom, soft bodies pose unique challenges in terms of simulation, control, and optimization. Herein I propose a novel soft-bodied agents formalism, namely, pressure-based soft agents (PSAs): spring-mass membranes containing a pressurized medium. Pressure endows the agents with structure, while springs and masses simulate softness and allow the agents to assume a large gamut of shapes. PSAs actuate both locally, by changing the resting lengths of springs, and globally, by modulating global pressure. I optimize the controller of PSAs for a locomotion task on hilly terrain, an escape task from a cage, and an object manipulation task. The results suggest that PSAs are indeed effective at the tasks, especially those requiring a shape change. I envision PSAs as playing a role in modeling soft-bodied agents, such as soft robots and biological cells. [ABSTRACT FROM AUTHOR]

This article addresses the co-evolution of morphology and control in evolutionary robotics, focusing on the challenge of premature convergence and limited morphological diversity. We conduct a comparative analysis of state-of-the-art algorithms, focusing on QD (Quality-Diversity) algorithms, based on a well-defined methodology for benchmarking evolutionary algorithms. We introduce carefully chosen indicators to evaluate their performance in three core aspects: task performance, phenotype diversity, and genotype diversity. Our findings highlight MNSLC (Multi-BC NSLC), with the introduction of aligned novelty to NSLC (Novelty Search with Local Competition), as the most effective algorithm for diversity preservation (genotype and phenotype diversity), while maintaining a competitive level of exploitability (task performance). MAP-Elites, although exhibiting a well-balanced trade-off between exploitation and exploration, fall short in protecting morphological diversity. NSLC, while showing similar performance to MNSLC in terms of exploration, is the least performant in terms of exploitation, contrasting with QN (Fitness-Novelty MOEA), which exhibits much superior exploitation, but inferior exploration, highlighting the effects of local competition in skewing the balance toward exploration. Our study provides valuable insights into the advantages, disadvantages, and trade-offs of different algorithms in co-evolving morphology and control. [ABSTRACT FROM AUTHOR]

The discrepancy between simulated and hardware experiments, the reality gap, is a challenge in evolutionary robotics. While strategies have been proposed to address this gap in fixed-body robots, they are not viable when dealing with populations and generations where the body is in constant change. The continual evolution of body designs necessitates the manufacturing of new robotic structures, a process that can be time-consuming if carried out manually. Moreover, the increased manufacturing time not only prolongs hardware experimental durations but also disrupts the synergy between hardware and simulated experiments. Failure to effectively manage these challenges could impede the implementation of evolutionary robotics in real-life environments. The Autonomous Robot Evolution project presents a framework to tackle these challenges through a case study. This paper describes the main three contributions of this work: Firstly, it analyses the different reality gap experienced by each different robot or the heterogenous reality gap. Secondly, it emphasizes the importance of automation in robot manufacturing. And thirdly, it highlights the necessity of a framework to orchestrate the synergy between simulated and hardware experiments. In the long term, integrating these contributions into evolutionary robotics is envisioned to enable the continuous production of robots in real-world environments.

A swarm of robots can collectively select an option among the available alternatives offered by the environment through a process known as collective decision-making. This process is characterised by the fact that once the group makes a decision, it can not be attributed to any of its group members. In swarm robotics, the individual mechanisms for collective decision-making are generally hand-designed and limited to a restricted set of solutions based on the voter or the majority model. In this paper, we demonstrate that it is possible to take an alternative approach in which the individual mechanisms are implemented using artificial neural network controllers automatically synthesised using evolutionary computation techniques. We qualitatively describe the group dynamics underlying the decision process on a collective perceptual discrimination task. We carry out extensive comparative tests that quantitatively evaluate the performance of the most commonly used decision-making mechanisms (voter model and majority model) with the proposed dynamic neural network model under various operating conditions and for swarms that differ in size. The results of our study clearly indicate that the performances of a swarm employing dynamical neural networks as the decision-making mechanism are more robust, more adaptable to a dynamic environment, and more scalable to a larger swarm size than the performances of the swarms employing the voter and the majority model. These results, generated in simulation, are ecologically validated on a swarm of physical e-puck2 robots. [ABSTRACT FROM AUTHOR]

The increasing number of robots and the rising cost of electricity have spurred research into energy-reducing concepts in robotics. One such concept, elastic actuation, introduces compliant elements such as springs into the robot structure. This article presents a comparative analysis between two types of elastic actuation, namely, monoarticular parallel elastic actuation and biarticular parallel elastic actuation, and demonstrates an end-to-end pipeline for their optimization. Starting from the real-world system identification of an RRR robotic arm, we calibrate a simulation model in a general-purpose physics engine and employ in silico evolutionary optimization to co-optimize spring configurations and trajectories for a pick-and-place task. Finally, we successfully transfer the in silico optimized elastic elements and trajectory to the real-world prototype. Our results substantiate the ability of elastic actuation to reduce the actuators' torque requirements heavily. In contrast to previous work, we highlight the superior performance of the biarticular variant over the monoarticular configuration. Furthermore, we show that a combination of both proves most effective. This work provides valuable insights into the torque-reducing use of elastic actuation and demonstrates an actuator-invariant in silico optimization methodology capable of bridging the sim2real gap. [ABSTRACT FROM AUTHOR]

Swarm robotics is a distinctive type of multi-robotic system that relies on local communication among the swarm members to generate a desired global behaviour. This implies a lack of global information, requiring robots to sense and communicate using sensors and actuators located on their bodies. Consequently, the robots within the swarm must leverage collective intelligence to solve the problem at hand, as no individual robot can accomplish the task independently. This article provides an overview of swarm robotics in general, highlighting its characteristics that distinguish it from other multi-robotic systems and simultaneously serve as motivation to adopt a swarm robotics approach. A closer examination of collective decision making within swarm robotics and its design problem also provided, classifying design methods into manual design and automatic approaches. The most commonly used automatic approaches to design collective decision making in swarm robotics are explained, along with a mention of the benefits and drawbacks of such approaches. However, this review does not cover aspects such as the swarm collective behaviours – except collective decision making – and the swarm robotics tasks. [ABSTRACT FROM AUTHOR]

Development of computational techniques, robotics, 3D printing and materials technologies enables the creation of advanced robotic systems that can autonomously reproduce and evolve. Emerging robotic evolution technology challenges existing AI ethics. The inherent adaptability, stochasticity and complexity of evolutionary systems pose numerous threats. It is difficult to underestimate the possible implications of two key functionalities of evolving robots: self-replication and random change of robot form and behavior. Self-replication allows robots to reproduce without human intervention. Mutations or random evolutionary changes can lead to undesirable and harmful behavior of robots, threatening human interests. Whenever a technology is developed that is not under direct human control and the process is unpredictable, questions of risk and liability arise. The article addresses both possible risk of harm and liability in connection with the key issue of human control over the process of evolution. The issue of responsibility for artificial intelligence was considered to be particularly important from both an ethical and legal point of view. In general, responsibility covers the full spectrum of events ex post (who is at fault, what was the reason), and ex ante (what actions should be taken to reduce the risk, i.e. what is negligence and who commits it), and usually rests with people. The problem, however, is the existence of gaps in the scope of liability for damages or adverse events involving people and artificial intelligence systems that cannot be filled with traditional concepts of assigning responsibility. The article indicates a method of comprehensively solving the problem of gaps in liability for artificial intelligence, proposed in the literature, based on the idea of designing social engineering systems that enable “meaningful human control”, i.e. systems adapted to human intentions and capabilities. The aim of the article is to indicate the urgent need for a continuous search for an answer to the question of how to responsibly control the evolution of robots.

This paper demonstrates a controller design of a multi-legged robotic swarm in a rough terrain environment. Many studies in swarm robotics are conducted with mobile robots that work in relatively flat fields. This paper focuses on a multi-legged robotic swarm, which is expected to operate not only in a flat field but also in rough terrain environments. However, designing a robot controller becomes a challenging problem because a designer has to consider how to coordinate a large number of joints in a robot, besides the complexity of a swarm problem. This paper employed an evolutionary robotics approach for the automatic design of a robot controller. The experiments were conducted by computer simulations with the path formation task. The results showed that the proposed approach succeeds in generating collective behavior in flat and rough terrain environments. [ABSTRACT FROM AUTHOR]

This paper focuses on generating collective behavior of a robotic swarm using an attention agent. The selective attention mechanism enables an agent to cope with environmental variations which are irrelevant to the task. This paper applies attention mechanisms to a robotic swarm for enhancing system-level properties, such as flexibility or scalability. To train an attention agent, evolutionary computations become a promising method, because a controller structure is not restricted by a gradient-based method. Therefore, this paper employs a deep neuroevolution approach to generating collective behavior in a robotic swarm. The experiments are conducted by computer simulations that consist of the Unity 3D game engine. The performance of the attention agent is compared with the convolutional neural network approach. The experimental results showed that the attention agent obtained generalization abilities in a robotic swarm similar to single-agent problems. [ABSTRACT FROM AUTHOR]

While evolutionary robotics can create novel morphologies and controllers that are well-adapted to their environments, learning is still the most efficient way to adapt to changes that occur on shorter time scales. Learning proposals for evolving robots to date have focused on new individuals either learning a controller from scratch, or building on the experience of direct ancestors and/or robots with similar configurations. Here we propose and demonstrate a novel means for social learning of gait patterns, based on sensorimotor synchronization. Using movement patterns of other robots as input can drive nonlinear decentralized controllers such as CPGs into new limit cycles, hence encouraging diversity of movement patterns. Stable autonomous controllers can then be locked in, which we demonstrate using a quasi-Hebbian feedback scheme. We propose that in an ecosystem of robots evolving in a heterogeneous environment, such a scheme may allow for the emergence of generalist task-solvers from a population of specialists.

Embodied evolution is an evolutionary robotics approach that implements an evolutionary algorithm over a population of robots and evolves while the robots perform their tasks. In embodied evolution, robots send and receive genomes from their neighbors and generate an offspring genome from the exchanged genomes. This study focused on the effects of the communication range for exchanging genomes on the evolvability of embodied evolution. Experiments were conducted using computer simulations, where robot controllers were evolved during a two-target navigation task. The results of the experiments showed that the robotic swarm could achieve better performance by reducing the communication range for exchanging genomes. [ABSTRACT FROM AUTHOR]

Swarm robotics system (SRS) is a type of artifact that employs multiple robots to work together in a coordinated way, inspired by the self-organizing behavior of social insects such as ants and bees. SRSs are known for their robustness, flexibility, and scalability. This study focuses on evolutionary robotics (ER) which uses artificial neural networks (ANNs) as controllers to operate autonomous robots. In traditional ER research, SRSs were often composed of teams of homogeneous robots, each of which is controlled by a single ANN. In contrast, this study focuses on the implementation of ER in a heterogeneous SRS. To evaluate our approach, we present the concept of employing multiple controllers for sub-teams in a swarm. Heterogeneity was achieved using different controllers for the same physical bodies. We simulated a cooperative transport task, in which the performance of heterogeneity was superior because the two ANN controllers were able to express a variety of behaviors as an entire swarm. Additionally, this study investigated how well the three types of parental selection methods of the heterogeneous approach, can help to optimize the performance of the swarm. [ABSTRACT FROM AUTHOR]

In this research, a biped robot is programmed to learn to walk using genetic algorithms. The biped robot has the ability to rise after a fall. The system is made according to human genetics, where a child tries to learn to walk, takes a few steps, falls, and then stands up somehow after a fall. After this process, the system restarts and tries to walk again. The biped robot evolves after each generation, and after each generation it will have a more natural walking posture. The system is unpredictable, because the biped robot sometimes evolves faster and sometimes evolves slower, like in human genetics where some children are faster and other children are slower to walk. The biped robot is considered to have evolved enough after having a walking posture similar to that of a human and does not fall. [ABSTRACT FROM AUTHOR]

The aim of this paper is to present the genetic algorithm used in programming a solar tracker robot. The solar tracker robot can be used to increase the efficiency of solar panels by rotating them toward the Sun. When the efficiency of solar panels is increased, they can reduce the space occupied by them. Efficiency can also be increased by genetic algorithms, where a solar panel can evolve and thus track the Sun more precisely. The genetic algorithm implemented in the solar tracker is the main focus of this paper. The genetic algorithm is presented with a flow chart and formulas. After this, the system is implemented on a solar tracker robot and validated with real experiments. The solar tracker robot algorithm is implemented in the C language on a microcontroller built on an FPGA platform. [ABSTRACT FROM AUTHOR]

The evolutionary robotics field offers the possibility of autonomously generating robots that are adapted to desired tasks by iteratively optimising across successive generations of robots with varying configurations until a high-performing candidate is found. The prohibitive time and cost of actually building this many robots means that most evolutionary robotics work is conducted in simulation, but to apply evolved robots to real-world problems, they must be implemented in hardware, which brings new challenges. This paper explores in detail the design of an example system for realising diverse evolved robot bodies, and specifically how this interacts with the evolutionary process. We discover that every aspect of the hardware implementation introduces constraints that change the evolutionary space, and exploring this interplay between hardware constraints and evolution is the key contribution of this paper. In simulation, any robot that can be defined by a suitable genetic representation can be implemented and evaluated, but in hardware, real-world limitations like manufacturing/assembly constraints and electrical power delivery mean that many of these robots cannot be built, or will malfunction in operation. This presents the novel challenge of how to constrain an evolutionary process within the space of evolvable phenotypes to only those regions that are practically feasible: the viable phenotype space. Methods of phenotype filtering and repair were introduced to address this, and found to degrade the diversity of the robot population and impede traversal of the exploration space. Furthermore, the degrees of freedom permitted by the hardware constraints were found to be poorly matched to the types of morphological variation that would be the most useful in the target environment. Consequently, the ability of the evolutionary process to generate robots with effective adaptations was greatly reduced. The conclusions from this are twofold. 1) Designing a hardware platform for evolving robots requires different thinking, in which all design decisions should be made with reference to their impact on the viable phenotype space. 2) It is insufficient to just evolve robots in simulation without detailed consideration of how they will be implemented in hardware, because the hardware constraints have a profound impact on the evolutionary space.

Embodied evolution is an evolutionary robotics approach that implements an evolutionary algorithm over a population of robots and evolves while the robots perform their tasks. So far, most studies on embodied evolution utilize relatively simple neural networks as robot controllers. However, a simple structured controller might restrict robot behavior and lead to lower performance. This paper proposes an embodied evolution approach that uses echo state networks as robot controllers. The experiments are conducted using computer simulations, and the controllers are evolved in a two-target navigation task. The results show that the echo state network controllers outperform the conventional controllers. [ABSTRACT FROM AUTHOR]

The technological revolution in the fields of robotics and artificial intelligence seems to indicate a future shift in our human-centered social paradigm towards a greater inclusion of artificial cognitive agents in our everyday environments. Understanding AI agents goes beyond the interpretation of a specific algorithm and requires analyzing the interactions between agents and with their surroundings. Combining AI and behavioral science is then seen in its true potential, as an ethical way to improve people's lives, where end users can feel safe when sharing personal data for transparent and relevant services.

Due to the lack of systematic empirical analyses and comparisons of ideas and methods, a clearly established state of the art is still missing in the optimization-based design of robot swarms. In this article, we propose an experimental protocol for the comparison of fully automatic design methods. This protocol is characterized by two notable elements: 1) a way to define benchmarks for the evaluation and comparison of design methods and 2) a sampling strategy that minimizes the variance when estimating their expected performance. To define generally applicable benchmarks, we introduce the notion of mission generator: a tool to generate missions that mimic those a design method will eventually have to solve. To minimize the variance of the performance estimation, we show that, under some common assumptions, one should adopt the sampling strategy that maximizes the number of missions considered—a formal proof is provided as the supplementary material. We illustrate the experimental protocol by comparing the performance of two offline fully automatic design methods that were presented in previous publications. [ABSTRACT FROM AUTHOR]

This paper demonstrates to generate a collective behavior of a multi-legged robotic swarm based on the evolutionary robotics approach. Most studies in swarm robotics are conducted using mobile robots driven by wheels. This paper focuses on generating collective behavior using a multi-legged robotic swarm. The evolutionary robotics approach is employed for designing a robot controller. The intuition-based constraint factors are incorporated into the fitness function to make the gait of robots similar to natural organisms. The experiment on a task of forming a line is conducted in computer simulations using the PyBullet physics engine. The robot controller is represented by a recurrent neural network with a single hidden layer. The experimental results show that proposed constraint factors successfully designed the robot's gait similar to natural organisms. The results also show that the evolutionary robotics approach successfully designed the robot controller for collective behavior of a multi-legged robotic swarm. [ABSTRACT FROM AUTHOR]

In Evolutionary Robotics, Lexicase selection has proven effective when a single task is broken down into many individual parameterizations. Evolved individuals have generalized across unique configurations of an overarching task. Here, we investigate the ability of Lexicase selection to generalize across multiple tasks, with each task again broken down into many instances. There are three objectives: to determine the feasibility of introducing additional tasks to the existing platform; to investigate any consequential effects of introducing these additional tasks during evolutionary adaptation; and to explore whether the schedule of presentation of the additional tasks over evolutionary time affects the final outcome. To address these aims we use a quadruped animat controlled by a feed-forward neural network with joint-angle, bearing-to-target, and spontaneous sinusoidal inputs. Weights in this network are trained using evolution with Lexicase-based parent selection. Simultaneous adaptation in a wall crossing task (labelled wall-cross) is explored when one of two different alternative tasks is also present: turn-and-seek or cargo-carry. Each task is parameterized into 100 distinct variants, and these variants are used as environments for evaluation and selection with Lexicase. We use performance in a single-task wall-cross environment as a baseline against which to examine the multi-task configurations. In addition, the objective sampling strategy (the manner in which tasks are presented over evolutionary time) is varied, and so data for treatments implementing uniform sampling, even sampling, or degrees of generational sampling are also presented. The Lexicase mechanism successfully integrates evolution of both turn-and-seek and cargo-carry with wall-cross, though there is a performance penalty compared to single task evolution. The size of the penalty depends on the similarity of the tasks. Complementary tasks (wallcross/turn-and-seek) show better performance than antagonistic tasks (wall-cross/cargo-carry). In complementary tasks performance is not affected by the sampling strategy. Where tasks are antagonistic, uniform and even sampling strategies yield significantly better performance than generational sampling. In all cases the generational sampling requires more evaluations and consequently more computational resources. The results indicate that Lexicase is a viable mechanism for multitask evolution of animat neurocontrollers, though the degree of interference between tasks is a key consideration. The results also support the conclusion that the naive, uniform random sampling strategy is the best choice when considering final task performance, simplicity of implementation, and computational efficiency. [ABSTRACT FROM AUTHOR]

This paper deals with the topology and weight evolution of Neuro-Controllers (NCs). It focuses on meta-problems of developing motion controllers that can operate successfully in several given motion-problems, where each of these motion problems defers by the arena and task. Here, the evolution aims to find both specialized and non-specialized controllers. The non-specialized controllers are evolved to provide a successful motion for the entire set of the given motion-problems, whereas each specialized controller is evolved to be optimal in at least one of these problems. The meta-problem is defined as a many-objective optimization problem, in which the kth objective is to maximize the motion performance in the kth motion problem. Following the problem description, a decomposition-based evolutionary algorithm, which is termed NEWS/D, is presented. This algorithm, which has recently been proposed by the authors, is specially designed to allow the topology and weight evolution of NCs under a large number of objectives. Next, the algorithm is applied to find NCs for three meta-problems, i.e., three sets of motion problems. The results of the experiments show that the obtained three sets of NCs include both specialized and non-specialized controllers. In addition, the proposed approach of many-objective topology and weight evolution of NCs is analyzed in comparison with other approaches, including three versions of decomposition-based fixed topology optimization. It is shown here that for the same number of evaluations, NEWS/D achieved better results than those of the other approaches. Finally, a topology analysis is carried out for the specialized NCs, which suggests that the solution includes multiple equivalent NCs. The numerical implications of this phenomenon are left for future work. [ABSTRACT FROM AUTHOR]

Quality-diversity (QD) algorithms evolve behaviorally diverse and high-performing solutions. To illuminate the elite solutions for a space of behaviors, QD algorithms require the definition of a suitable behavior space. If the behavior space is high-dimensional, a suitable dimensionality reduction technique is required to maintain a limited number of behavioral niches. While current methodologies for automated behavior spaces focus on changing the geometry of the behavior space or on unsupervised learning of its key features, there remains a need for customizing behavioral diversity to a particular meta-objective specified by the end user. In the newly emerging framework of QD Meta-Evolution, or QD-Meta for short, one evolves a population of QD algorithms, each with different algorithmic and representational characteristics, to optimize the algorithms and their resulting archives to a user-defined meta-objective. Despite promising results compared to traditional QD algorithms, QD-Meta has yet to be compared to state-of-the-art behavior space automation methods, such as centroidal voronoi tessellations multidimensional archive of phenotypic elites (CVT-MAP-Elites) and autonomous robots realizing their abilities (AURORAs). This article performs an empirical study of QD-Meta on function optimization and multilegged robot locomotion benchmarks. Results demonstrate that QD-Meta archives provide improved average performance and faster adaptation to a priori unknown changes to the environment when compared to CVT-MAP-Elites and AURORA. A qualitative analysis shows how the resulting archives are tailored to the meta-objectives provided by the end user. [ABSTRACT FROM AUTHOR]

Living systems process sensory data to facilitate adaptive behavior. A given sensor can be stimulated as the result of internally driven activity, or by purely external (environmental) sources. It is clear that these inputs are processed differently—have you ever tried tickling yourself? Self-caused stimuli have been shown to be attenuated compared to externally caused stimuli. A classical explanation of this effect is that when the brain sends a signal that would result in motor activity, it uses a copy of that signal to predict the sensory consequences of the resulting motor activity. The predicted sensory input is then subtracted from the actual sensory input, resulting in attenuation of the stimuli. To critically evaluate the utility of this predictive approach for coping with self-caused stimuli, and investigate when non-predictive solutions may be viable, we implement a computational model of a simple embodied system with self-caused sensorimotor dynamics, and use a genetic algorithm to explore the solutions possible in this model. We find that in this simple system the solutions that emerge modify their behavior to shape or avoid self-caused sensory inputs, rather than predicting these self-caused inputs and filtering them out. In some cases, solutions take advantage of the presence of these self-caused inputs. The existence of these non-predictive solutions demonstrates that embodiment provides possibilities for coping with self-caused sensory interference without the need for an internal, predictive model.

The propensity of evolutionary algorithms to generate compact solutions have advantages and disadvantages. On one side, compact solutions can be cheaper, lighter, and faster than less compact ones. On the other hand, compact solutions might lack evolvability, i.e. might have a lower probability to improve as a result of genetic variations. In this work we study the relation between phenotypic complexity and evolvability in the case of soft-robots with varying morphology. We demonstrate a correlation between phenotypic complexity and evolvability. We demonstrate that the tendency to select compact solutions originates from the fact that the fittest robots often correspond to phenotypically simple robots which are robust to genetic variations but lack evolvability. Finally, we demonstrate that the efficacy of the evolutionary process can be improved by increasing the probability of genetic variations which produce a complexification of the agents’ phenotype or by using absolute mutation rates.