Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
225
result(s) for
"Evolutionary robotics Computer simulation."
Sort by:
Morphological change in machines accelerates the evolution of robust behavior
Most animals exhibit significant neurological and morphological change throughout their lifetime. No robots to date, however, grow new morphological structure while behaving. This is due to technological limitations but also because it is unclear that morphological change provides a benefit to the acquisition of robust behavior in machines. Here I show that in evolving populations of simulated robots, if robots grow from anguilliform into legged robots during their lifetime in the early stages of evolution, and the anguilliform body plan is gradually lost during later stages of evolution, gaits are evolved for the final, legged form of the robot more rapidly— and the evolved gaits are more robust— compared to evolving populations of legged robots that do not transition through the anguilliform body plan. This suggests that morphological change, as well as the evolution of development, are two important processes that improve the automatic generation of robust behaviors for machines. It also provides an experimental platform for investigating the relationship between the evolution of development and robust behavior in biological organisms.
Journal Article
A Comprehensive Review of Swarm Optimization Algorithms
Many swarm optimization algorithms have been introduced since the early 60's, Evolutionary Programming to the most recent, Grey Wolf Optimization. All of these algorithms have demonstrated their potential to solve many optimization problems. This paper provides an in-depth survey of well-known optimization algorithms. Selected algorithms are briefly explained and compared with each other comprehensively through experiments conducted using thirty well-known benchmark functions. Their advantages and disadvantages are also discussed. A number of statistical tests are then carried out to determine the significant performances. The results indicate the overall advantage of Differential Evolution (DE) and is closely followed by Particle Swarm Optimization (PSO), compared with other considered approaches.
Journal Article
From Bioinspiration to Computer Generation: Developments in Autonomous Soft Robot Design
The emerging field of soft robotics presents a new paradigm for robot design in which “precision through rigidity” is replaced by “cognition through compliance.” Lightweight and flexible, soft robots have vast potential to interact with fragile objects and navigate unstructured environments. Like octopuses and worms in nature, soft robots’ flexible bodies conform to hard objects and reconfigure for different tasks, delegating the burden of control from brain to body through embodied cognition. However, because of the lack of efficient modeling and simulation tools, soft robots are primarily designed by hand. Typically, hard components from rigid robots or living creatures are heuristically substituted for comparable soft ones. Autonomous design and manufacturing methodologies are urgently required to produce bespoke, high‐performing robots. Currently, design methodologies exist between simple but realistic parametric optimizations, and evolutionary algorithms which simulate morphology and control coevolution. To find high‐performing designs, novel high‐fidelity simulators and high‐throughput manufacturing and testing processes are required to explore the complex soft material, morphology and control landscape, blending simulation, and experimental data. This article reviews the state of the art in autonomous soft robotic design. Existing manual and automated designs are surveyed and future directions to automate soft robot design and manufacturing are presented.
By using soft and functional materials to deform around objects and adapt to new environments, soft robotics has the potential to revolutionize material handling and terrain navigation. But in the absence of accurate modeling tools, they are still laboriously designed manually. This article reviews progress toward autonomous modeling, simulation, and design.
Journal Article
The evolutionary origins of modularity
A central biological question is how natural organisms are so evolvable (capable of quickly adapting to new environments). A key driver of evolvability is the widespread modularity of biological networks—their organization as functional, sparsely connected subunits—but there is no consensus regarding why modularity itself evolved. Although most hypotheses assume indirect selection for evolvability, here we demonstrate that the ubiquitous, direct selection pressure to reduce the cost of connections between network nodes causes the emergence of modular networks. Computational evolution experiments with selection pressures to maximize network performance and minimize connection costs yield networks that are significantly more modular and more evolvable than control experiments that only select for performance. These results will catalyse research in numerous disciplines, such as neuroscience and genetics, and enhance our ability to harness evolution for engineering purposes.
Journal Article
Semi-supervised semantic segmentation in Earth Observation: the MiniFrance suite, dataset analysis and multi-task network study
The development of semi-supervised learning techniques is essential to enhance the generalization capacities of machine learning algorithms. Indeed, raw image data are abundant while labels are scarce, therefore it is crucial to leverage unlabeled inputs to build better models. The availability of large databases have been key for the development of learning algorithms with high level performance. Despite the major role of machine learning in Earth Observation to derive products such as land cover maps, datasets in the field are still limited, either because of modest surface coverage, lack of variety of scenes or restricted classes to identify. We introduce a novel large-scale dataset for semi-supervised semantic segmentation in Earth Observation, the MiniFrance suite. MiniFrance has several unprecedented properties: it is large-scale, containing over 2000 very high resolution aerial images, accounting for more than 200 billions samples (pixels); it is varied, covering 16 conurbations in France, with various climates, different landscapes, and urban as well as countryside scenes; and it is challenging, considering land use classes with high-level semantics. Nevertheless, the most distinctive quality of MiniFrance is being the only dataset in the field especially designed for semi-supervised learning: it contains labeled and unlabeled images in its training partition, which reproduces a life-like scenario. Along with this dataset, we present tools for data representativeness analysis in terms of appearance similarity and a thorough study of MiniFrance data, demonstrating that it is suitable for learning and generalizes well in a semi-supervised setting. Finally, we present semi-supervised deep architectures based on multi-task learning and the first experiments on MiniFrance. These results will serve as baselines for future work on semi-supervised learning over the MiniFrance dataset. The Minifrance suite and related semi-supervised networks will be publicly available to promote semi-supervised works in Earth Observation.
Journal Article
Model parameterization of robotic systems through the bio-inspired optimization
The accurate modeling of dynamic systems, particularly robotic ones, is crucial in the industry. It enables simulation-based approaches that facilitate various tasks without requiring the physical system, thereby reducing risks and costs. These approaches range from model-in-the-loop (MiL), where a simulated model of the real plant is used for controller design, to hardware-in-the-loop (HiL), which provides more realistic simulations on specialized real-time hardware. Among these, MiL is widely adopted due to its simplicity and effectiveness in developing control strategies. However, to fully leverage the advantages of MiL, developing a robust and accurate system model parameterization methodology is essential. This methodology should be adaptable to a wide range of applications, adopt a holistic approach, and balance the cost-benefit trade-offs in model characteristics. Achieving this, however, introduces additional challenges related to system complexity and the inherent properties of the model. To address these challenges, this work proposes a model parameterization approach for robotic systems using bio-inspired optimization to develop accurate and practical models for system design. The approach formulates an optimization problem to determine the dynamic model parameters of a robot, ensuring its behavior closely resembles that of the real system. Due to the complexity of this problem, bio-inspired optimization techniques are particularly well-suited. The proposed method is validated using a theoretical, non-conservative model of a three-degree-of-freedom serial robot. The dynamic parameters of its three links were identified to effectively generalize the real system. To solve the optimization problem, three bio-inspired algorithms were employed: the genetic algorithm, particle swarm optimization, and differential evolution. The optimal parameterization obtained for the robot model demonstrated the effectiveness of the proposed approach in a MiL simulation environment, achieving an overall correlation of 0.9019 in the experiments. This correlation highlights the model’s ability to predict the robot’s behavior accurately. Additionally, the methodology’s efficacy was further validated in another electromechanical system, the reaction force-sensing series elastic actuator, yielding a correlation of 0.8379 in the resulting model.
Journal Article
A new prediction model of battery and wind-solar output in hybrid power system
In this paper short term power forecast of wind and solar power is proposed to evaluate the available output power of each production component. In this model, lead acid batteries used in proposed hybrid power system based on wind-solar power system. So, before the predicting of power output, a simple mathematical approach to simulate the lead–acid battery behaviors in stand-alone hybrid wind-solar power generation systems will be introduced. Then, the proposed forecast problem will be evaluated which is taken as constraint status through state of charge (SOC) of the batteries. The proposed forecast model includes a feature selection filter and hybrid forecast engine based on neural network (NN) and an intelligent evolutionary algorithm. This method not only could maintain the SOC of batteries in suitable range, but also could decrease the on-or-off switching number of wind turbines and PV modules. Effectiveness of the proposed method has been applied over real world engineering data. Obtained numerical analysis, demonstrate the validity of proposed method.
Journal Article
Tiles: an online algorithm for community discovery in dynamic social networks
Community discovery has emerged during the last decade as one of the most challenging problems in social network analysis. Many algorithms have been proposed to find communities on static networks, i.e. networks which do not change in time. However, social networks are dynamic realities (e.g. call graphs, online social networks): in such scenarios static community discovery fails to identify a partition of the graph that is semantically consistent with the temporal information expressed by the data. In this work we propose
Tiles
, an algorithm that extracts overlapping communities and tracks their evolution in time following an online iterative procedure. Our algorithm operates following a domino effect strategy, dynamically recomputing nodes community memberships whenever a new interaction takes place. We compare
Tiles
with state-of-the-art community detection algorithms on both synthetic and real world networks having annotated community structure: our experiments show that the proposed approach is able to guarantee lower execution times and better correspondence with the ground truth communities than its competitors. Moreover, we illustrate the specifics of the proposed approach by discussing the properties of identified communities it is able to identify.
Journal Article
Maximal coverage problems with routing constraints using cross-entropy Monte Carlo tree search
Spatial search, and environmental monitoring are key technologies in robotics. These problems can be reformulated as maximal coverage problems with routing constraints, which are NP-hard problems. The generalized cost-benefit algorithm (GCB) can solve these problems with theoretical guarantees. To achieve better performance, evolutionary algorithms (EA) boost its performance via more samples. However, it is hard to know the terminal conditions of EA to outperform GCB. To solve these problems with theoretical guarantees and terminal conditions, in this research, the cross-entropy based Monte Carlo Tree Search algorithm (CE-MCTS) is proposed. It consists of three parts: the EA for sampling the branches, the upper confidence bound policy for selections, and the estimation of distribution algorithm for simulations. The experiments demonstrate that the CE-MCTS outperforms benchmark approaches (e.g., GCB, EAMC) in spatial search problems.
Journal Article