Catálogo de publicaciones - libros

Embodied artificial intelligence is based on the notion that cognition and action emerge from interactions between brain, body and environment. This chapter sketches a set of foundational principles that might be useful for understanding the emergence (“discovery”) of intelligence in biological and artificial embodied systems. Special emphasis is placed on information as a crucial resource for organisms and on information theory as a promising descriptive and predictive framework linking morphology, perception, action and neural control.

This paper exposes and discusses the concept of ’networked embodied cognition’, based on natural embodied neural networks, with some considerations on the nature of natural collective intelligence and cognition, and with reference to natural biological examples, evolution theory, neural network science and technology results, network robotics. It shows that this could be the method of cognitive adaptation to the environment most widely used by living systems and most fit to the deployment of artificial robotic networks. Some preliminary ideas about the development of a quantitative framework are shortly discussed. On the basis of the work of many people a few approximate simple quantitative relations are derived between information metrics of the phase space behavior of the agent dynamical system and those of the cognition system perceived by an external observer.

In this paper we have investigated a group of multi agent systems (MAS) in which the agents change their environment and this change has the potential to trigger behaviors in other agents of the group in another time or another position in the environment. The structure makes it possible to conceptualize the group as a super organism incorporating the agents and the environment such that new behaviors are observed from the whole group as a result of the specific distribution of agents in that environment. This distribution exists in many aspects like a super memory (or even a super brain) that exists in the environment and is not limited to memories of the individuals. There is a distributed decision making that is done by the group of agents which, in a higher level consists of both individual and group decision makings, and can be viewed as emergent rather than consciously planned. As the agents change the environment, they decrease the error for the group and hence a distributed learning is also forming between the agents of the group. This implicit learning is related to the implicit memory existing in the environment. These two interpretations of memory and learning are assessed with experimental results where two robots perform a task while they are not aware of their global behavior.

This article discusses the issues of adaptive autonomous navigation as a challenge of artificial intelligence. We argue that, in order to enhance the dexterity and adaptivity in robot navigation, we need to take into account the decentralized mechanisms which exploit physical system-environment interactions. In this paper, by introducing a few underactuated locomotion systems, we explain (1) how mechanical body structures are related to motor control in locomotion behavior, (2) how a simple computational control process can generate complex locomotion behavior, and (3) how a motor control architecture can exploit the body dynamics through a learning process. Based on the case studies, we discuss the challenges and perspectives toward a new framework of adaptive robot control.

This paper introduces our robotic case study which is intended to intensively investigate the neural-body coupling, , how the task distribution between control and mechanical systems should be achieved, so as to emerge useful functionalities. One of the significant features of this case study is that we have employed a collective behavioral approach. More specifically, we have focused on an “embodied” coupled nonlinear oscillator system by which we have generated one of the most primitive yet flexible locomotion, , amoeboid locomotion, in the hope that this primitiveness allows us to investigate the neural-body coupling effectively. Experiments we have conducted strongly support that there exists an “ecologically-balanced” task distribution, under which significant abilities such as real-time adaptivity emerge.

A new approach of integrating biological microorganisms such as bacteria to an inorganic robot body for propulsion in low velocity or stagnant flow field is proposed in this paper with the ultimate goal of fabricating a few hundreds of micrometer size swimming robots. To show the feasibility of this approach, bacteria are attached to microscale objects such as 10 micron polystyrene beads by blotting them in a bacteria swarm plate. Randomly attached bacteria are shown to propel the beads at an average speed of approximately 15 μm/sec stochastically. Using chemical stimuli, bacteria flagellar propulsion is halted by introducing copper ions into the motility medium of the beads, while ethylenediaminetetraacetic acid is used to resume their motion. Thus, repeatable on/off motion control of the bacteria integrated mobile beads was shown. On-board chemical motion control, steering, wireless communication, sensing, and position detection are few of the future challenges for this work. Small or large numbers of these microrobots can potentially enable hardware platforms for self-organization, swarm intelligence, distributed control, and reconfigurable systems in the future.

Compressing real-time input through bandwidth constrained connections has been studied within robotics, wireless sensor networks, and image processing. When there are bandwidth constraints on real-time input the amount of information to be transferred will always be greater than the amount that can be transferred per unit of time. We propose a system that utilizes a local diffusion process and a reinforcement learning-based memory system to establish a real-time prediction of an entire input space based upon partial observation. The proposed system is optimized for dealing with multi-dimension input spaces, and maintains the ability to react to rare events. Results show the relation of loss to quality and suggest that at higher resolutions gains in quality are possible.

The human brain is the best example of intelligence known, with unsurpassed ability for complex, real-time interaction with a dynamic world. AI researchers trying to imitate its remarkable functionality will benefit by learning more about neuroscience, and the differences between Natural and Artificial Intelligence. Steps that will allow AI researchers to pursue a more brain-inspired approach to AI are presented. A new approach that bridges AI and neuroscience is described, Embodied Cultured Networks. Hybrids of living neural tissue and robots, called hybrots, allow detailed investigation of neural network mechanisms that may inform future AI. The field of neuroscience will also benefit tremendously from advances in AI, to deal with their massive knowledge bases and help understand Natural Intelligence.

This case study demonstrates how the synthesis and the analysis of minimal recurrent neural robot control provide insights into the exploration of embodiment. By using structural evolution, minimal recurrent neural networks of general type were evolved for behavior control. The small size of the neural structures facilitates thorough investigations of behavior relevant neural dynamics and how they relate to interactions of robots within the sensorimotor loop. We argue that a clarification of dynamical neural control mechanisms in a reasonable depth allows quantitative statements about the effects of the sensorimotor loop and suggests general qualitative implications about the embodiment of autonomous robots and biological systems as well.

It is claimed that synaptic plasticity of neural controllers for autonomous robots can enhance the behavioral properties of these systems. Based on homeostatic properties of so called self-regulating neurons, the presented mechanism will vary the synaptic strength during the robot interaction with the environment, due to driving sensor inputs and motor outputs. This is exemplarily shown for an obstacle avoidance behavior in simulation.