This thesis presents two novel algorithms for the evolutionary optimisation of agent populations through divergent search of low dimensional phenotypic landscapes. As the eld of Evolutionary Robotics (ER) develops towards more complex domains, which often involve deception and uncertainty, the promotion of phenotypic diversity has become of increasing interest. Divergent exploration of the phenotypic feature space has been shown to avoid convergence towards local optima and to provide diverse sets of solutions to a given objective. Novelty Search (NS) and the more recent Multi-dimensional Archive of Phenotypic Elites (MAP-Elites), are two state of the art algorithms which utilise divergent phenotypic search. In this thesis, the individual merits and weaknesses of these algorithms are built upon in order to further develop the study of divergent phenotypic search within ER. An observation that the diverse range of individuals produced through the optimisation of novelty will likely contain solutions to multiple independent objectives is utilised to develop Multiple Assessment Directed Novelty Search (MADNS). The MADNS algorithm is introduced as an extension to NS for the simultaneous optimisation of multiple independent objectives, and is shown to become more e ective than NS as the size of the state space increases. The central contribution of this thesis is the introduction of a novel algorithm for rapid and thorough divergent search of low dimensional phenotypic landscapes. The Spatial, Hierarchical, Illuminated NeuroEvolution (SHINE) algorithm di ers from previous divergent search algorithms, in that it utilises a tree structure for the maintenance and selection of potential candidates. Unlike previous approaches, SHINE iteratively focusses upon sparsely visited areas of the phenotypic landscape without the computationally expensive distance comparison required by NS; rather, the sparseness of the area within the landscape where a potential solution resides is inferred through its depth within the tree. Experimental results in a range of domains show that SHINE signi cantly outperforms NS and MAP-Elites in both performance and exploration.

No description available.

The attempt to evolve complete embodied and situated artificial creatures in which both morphological and control characteristics are adapted during the evolutionary process has been and still represents a long term goal key for the artificial life and the evolutionary robotics community. Loosely inspired by ancient biological organisms which are not provided with a central nervous system and by simple organisms such as stick insects, this thesis proposes a new genotype encoding which allows development and evolution of mor- phology and neural controller in artificial agents provided with a distributed neural network. In order to understand if this kind of network is appropriate for the evolution of non trivial behaviours in artificial agents, two experiments (description and results will be shown in chapter 3) in which evolution was applied only to the controller’s parameters were performed. The results obtained in the first experiment demonstrated how distributed neural networks can achieve a good level of organization by synchronizing the output of oscillatory elements exploiting acceleration/deceleration mechanisms based on local interactions. In the second experiment few variants on the topology of neural architecture were introduced. Results showed how this new control system was able to coordinate the legs of a simulated hexapod robot on two different gaits on the basis of the external circumstances. After this preliminary and successful investigation, a new genotype encoding able to develop and evolve artificial agents with no fixed morphology and with a distributed neural controller was proposed. A second set of experiments was thus performed and the results obtained confirmed both the effectiveness of genotype encoding and the ability of distributed neural network to perform the given task. The results have also shown the strength of genotype both in generating a wide range of different morphological structures and in favouring a direct co-adaptation between neural controller and morphology during the evolutionary process. Furthermore the simplicity of the proposed model has showed the effective role of specific elements in evolutionary experiments. In particular it has demonstrated the importance of the environment and its complexity in evolving non-trivial behaviours and also how adding an independent component to the fitness function could help the evolutionary process exploring a larger space solutions avoiding a premature convergence towards suboptimal solutions.

The goal of the research activities presented in this thesis is the design of intelligent behaviours for a complex robotic system, which is composed of a swarm of autonomous units. Inspired by the organisational skills of social insects, we are particularly interested in the study of collective behaviours based on self-organisation. The problem of designing self-organising behaviours for a swarm of robots is tackled resorting to artificial evolution, which proceeds in a bottom-up direction by first defining the controllers at the individual level and then testing their effect at the collective level. In this way, it is possible to bypass the difficulties encountered in the decomposition of the global behaviour into individual ones, and the further encoding of the individual behaviours into the controllers' rules. In the experiments presented in this thesis, we show that this approach is viable, as it produces efficient individual controllers and robust self-organising behaviours. To the best of our knowledge, our experiments are the only example of evolved self-organising behaviours that are successfully tested on a physical robotic platform. Besides the engineering value, the evolution of self-organising behaviours for a swarm of robots also provides a mean for the understanding of those biological processes that were a fundamental source of inspiration in the first place. In this perspective, the experiments presented in this thesis can be considered an interesting instance of a synthetic approach to the study of collective intelligence and, more in general, of Cognitive Science.

From the very creation of the term by Czech writer Karel Capek in 1921, a "robot" has been synonymous with an artificial agent possessing a powerful body and cogitating mind. While the fields of Artificial Intelligence (AI) and Robotics have made progress into the creation of such an android, the goal of a cogitating robot remains firmly outside the reach of our technological capabilities. Cognition has proved to be far more complex than early AI practitioners envisioned. Current methods in Machine Learning have achieved remarkable successes in image categorization through the use of deep learning. However, when presented with novel or adversarial input, these methods can fail spectacularly. I postulate that a robot that is free to interact with objects should be capable of reducing spurious difference between objects of the same class. This thesis demonstrates and analyzes a robot that achieves more robust visual categorization when it first evolves to use proprioceptive sensors and is then trained to increasingly rely on vision, when compared to a robot that evolves with only visual sensors. My results suggest that embodied methods can scaffold the eventual achievement of robust visual classification.

Traditional techniques for the design of robots require human engineers to plan every aspect of the system, from body to controller. In contrast, the field of evolu- tionary robotics uses evolutionary algorithms to create optimized morphologies and neural controllers with minimal human intervention. In order to expand the capability of an evolved agent, it must be exposed to a variety of conditions and environments. This thesis investigates the design and benefits of virtual robots which can reflect the structure and modularity in the world around them. I show that when a robot’s morphology and controller enable it to perceive each environment as a collection of independent components, rather than a monolithic entity, evolution only needs to optimize on a subset of environments in order to maintain performance in the overall larger environmental space. I explore previously unused methods in evolutionary robotics to aid in the evolution of modularity, including using morphological and neurological cost. I utilize a tree morphology which makes my results generalizable to other mor- phologies while also allowing in depth theoretical analysis about the properties rel- evant to modularity in embodied agents. In order to better frame the question of modularity in an embodied context, I provide novel definitions of morphological and neurological modularity as well as create the sub-goal interference metric which mea- sures how much independence a robot exhibits with regards to environmental stimu- lus. My work extends beyond evolutionary robotics and can be applied to the opti- mization of embodied systems in general as well as provides insight into the evolution of form in biological organisms.

The goal of the research activities presented in this thesis is the design of intelligent behaviours for a complex robotic system, which is composed of a swarm of autonomous units. Inspired by the organisational skills of social insects, we are particularly interested in the study of collective behaviours based on self-organisation.<p><p>The problem of designing self-organising behaviours for a swarm of robots is tackled resorting to artificial evolution, which proceeds in a bottom-up direction by first defining the controllers at the individual level and then testing their effect at the collective level. In this way, it is possible to bypass the difficulties encountered in the decomposition of the global behaviour into individual ones, and the further encoding of the individual behaviours into the controllers' rules. In the experiments presented in this thesis, we show that this approach is viable, as it produces efficient individual controllers and robust self-organising behaviours. To the best of our knowledge, our experiments are the only example of evolved self-organising behaviours that are successfully tested on a physical robotic platform.<p><p>Besides the engineering value, the evolution of self-organising behaviours for a swarm of robots also provides a mean for the understanding of those biological processes that were a fundamental source of inspiration in the first place. In this perspective, the experiments presented in this thesis can be considered an interesting instance of a synthetic approach to the study of collective intelligence and, more in general, of Cognitive Science.<p> / Doctorat en sciences appliquées / info:eu-repo/semantics/nonPublished

Fundamental aspects of biologically-inspired regulatory mechanisms are considered in a robotics context, using artificial neural-network control systems . Regulatory mechanisms are used to control expression of genes, adaptation of form and behavior in organisms. Traditional neural network control architectures assume networks of neurons are fixed and are interconnected by wires. However, these architectures tend to be specified by a designer and are faced with several limitations that reduce scalability and tractability for tasks with larger search spaces. Traditional methods used to overcome these limitations with fixed network topologies are to provide more supervision by a designer. More supervision as shown does not guarantee improvement during training particularly when making incorrect assumptions for little known task domains. Biological organisms often do not require such external intervention (more supervision) and have self-organized through adaptation. Artificial neural tissues (ANT) addresses limitations with current neural-network architectures by modeling both wired interactions between neurons and wireless interactions through use of chemical diffusion fields. An evolutionary (Darwinian) selection process is used to ‘breed’ ANT controllers for a task at hand and the framework facilitates emergence of creative solutions since only a system goal function and a generic set of basis behaviours need be defined. Regulatory mechanisms are formed dynamically within ANT through superpositioning of chemical diffusion fields from multiple sources and are used to select neuronal groups. Regulation drives competition and cooperation among neuronal groups and results in areas of specialization forming within the tissue. These regulatory mechanisms are also shown to increase tractability without requiring more supervision using a new statistical theory developed to predict performance characteristics of fixed network topologies. Simulations also confirm the significance of regulatory mechanisms in solving certain tasks found intractable for fixed network topologies. The framework also shows general improvement in training performance against existing fixed-topology neural network controllers for several robotic and control tasks. ANT controllers evolved in a low-fidelity simulation environment have been demonstrated for a number of tasks on hardware using groups of mobile robots and have given insight into self-organizing system. Evidence of sparse activity and use of decentralized, distributed functionality within ANT controller solutions are found consistent with observations from neurobiology.

The works reported in this thesis focus upon synthesising neural controllers for anthropomorphic robots that are able to manipulate objects through an automatic design process based on artificial evolution. The use of Evolutionary Robotics makes it possible to reduce the characteristics and parameters specified by the designer to a minimum, and the robot’s skills evolve as it interacts with the environment. The primary objective of these experiments is to investigate whether neural controllers that are regulating the state of the motors on the basis of the current and previously experienced sensors (i.e. without relying on an inverse model) can enable the robots to solve such complex tasks. Another objective of these experiments is to investigate whether the Evolutionary Robotics approach can be successfully applied to scenarios that are significantly more complex than those to which it is typically applied (in terms of the complexity of the robot’s morphology, the size of the neural controller, and the complexity of the task). The obtained results indicate that skills such as reaching, grasping, and discriminating among objects can be accomplished without the need to learn precise inverse internal models of the arm/hand structure. This would also support the hypothesis that the human central nervous system (cns) does necessarily have internal models of the limbs (not excluding the fact that it might possess such models for other purposes), but can act by shifting the equilibrium points/cycles of the underlying musculoskeletal system. Consequently, the resulting controllers of such fundamental skills would be less complex. Thus, the learning of more complex behaviours will be easier to design because the underlying controller of the arm/hand structure is less complex. Moreover, the obtained results also show how evolved robots exploit sensory-motor coordination in order to accomplish their tasks.

Creating systems that can operate autonomously in complex environments is a challenge for contemporary engineering techniques. Automatic design methods offer a promising alternative, but so far they have not been able to produce agents that outperform manual designs. One such method is evolutionary robotics. It has been shown to be a robust and versatile tool for designing robots to perform simple tasks, but more challenging tasks at present remain out of reach of the method. In this thesis I discuss and attack some problems underlying the scalability issues associated with the method. I present a new technique for evolving modular networks. I show that the performance of modularity-biased evolution depends heavily on the morphology of the robot’s body and present a new method for co-evolving morphology and modular control. To be able to reason about the new technique I develop reformulation framework: a general way to describe and reason about metaoptimization approaches. Within this framework I describe a new heuristic for developing metaoptimization approaches that is based on the technique for co-evolving morphology and modularity. I validate the framework by applying it to a practical task of zero-g autonomous assembly of structures with a fleet of small robots. Although this work focuses on the evolutionary robotics, methods and approaches developed within it can be applied to optimization problems in any domain.