The interaction of an autonomous mobile robot with the real world critically depends on the robots morphology and on its environment. Building a model of these aspects is extremely complex, making simulation insufficient for accurate validation of control algorithms. If simulation environments are often very efficient, the tools for experimenting with real robots are often inadequate. The traditional programming languages and tools seldom provide enought support for realtime experiments, thus hindering the understanding of the control algorithms and making the experimentation complex and time-consuming. A miniature robot is presented: it has a cylindrical shape measuring 55 mm in diameter and 30 mm in height. Due to its small size, experiments can be performed quickly and cost-effectively in a small working area. Small peripherals can be designed and connected to the basic module and can take advantage of a versatile communication scheme. A serial-link is provided to run control algorithms on a workstation during debugging, thereby giving the user the opportunity of employing all available graphical tools. Once debugged, the algorithm can be downloaded to the robot and run on its own processor. Experimentation with groups of robots is hardly possible with commercially available hardware. The size and the price of the described robot open the way to cost-effective investigations into collective behaviour. This aspect of research drives the design of the robot described in this paper. Experiments with some twenty units are planned for the near future.

We have developed a methodology grounded in two beliefs: that autonomous agents need visual processing capabilities, and that the approach ofhand-designing control architectures for autonomous agents is likely to be superseded by methods involving the arti cial evolution of comparable architectures. In this paper we present results which demonstrate that neural-network control architectures can be evolved for an accurate simulation model of a visually guided robot. The simulation system involves detailed models of the physics of a real robot built at Sussex; and the simulated vision involves ray-tracing computer graphics, using models of optical systems which could readily be constructed from discrete components. The control-network architecture is entirely under genetic control, as are parameters governing the optical system. Signi cantly, we demonstrate that robust visually-guided control systems evolve from evaluation functions which do not explicitly involve monitoring visual input. The latter part of the paper discusses work now under development, which allows us to engage in long-term fundamental experiments aimed at thoroughly exploring the possibilities of concurrently evolving control networks and visual sensors for navigational tasks. This involves the construction of specialised visual-robotic equipment which eliminates the need for simulated sensing.

This report presents some experiments of a real-time navigation system driven by two cameras pointing laterally to the navigation direction (Divergent Stereo). Similarly to what has been proposed in [11; 5], our approach [17; 19] assumes that, for navigation purposes, the driving information is not distance (as it is obtainable by a stereo setup) but motion and, more precisely, by the use of qualitative optical flow information computed over nonoverlapping areas of the visual field of two cameras. Following this idea, a mobile vehicle has been equipped with a pair of cameras looking laterally (much like honeybees) and a controller based on fast, real-time computation of optical flow has been implemented. The control of the mobile robot (Robee) is based on the comparison between the apparent image velocity of the left and the right cameras. The solution adopted is derived from recent studies [21] describing the behavior of freely flying honeybees and the mechanisms they use to perceive ...

A qualitative approach to visually-guided navigation based on the computation of optical flow field is presented. The approach is based on the use of two cameras mounted on a mobile robot and with the optical axis directed in opposite directions such that the two visual fields do not overlap (divergent stereo); Range computation is based on the computation of the apparent image speed on images acquired during robot's motion. An example of reflex-type control of motion, driven by differential estimation of the flow field measured by the two eyes, is presented. In particular it is shown how a difficult task like navigating through a funneled corridor with obstacles, is possible without the need for metric depth estimation.

From perception to action and from action to perception, all elements of an autonomous agent are interdependent and need to be strongly coherent. The final behavior of the agent is the result of the global activity of this loop and every weakeness or incoherence of a single element has strong consequences on the performances of the agent. We think that, for the purpose of building autonomous robots, all these elements need to be developed together in continuous interaction with the environment. We describe the implementation of a possible solution (artificial neural networks and genetic algorithms) on a real mobile robot through a set of three different experiments. We focus our attention on three different aspects of the control structure: perception, internal representation and action. In all the experiments these aspects are not considered as single processing elements, but as part of an agent. For every experiment, the advantages and disadvantages of this approach are presented and...

Many arthropods (particularly insects) exhibit sophisticated visually guided behaviours. Yet in most cases the behaviours are guided by input from a few hundreds or thousands of "pixels " (i.e. ommatidia in the compound eye). Inspired by this observation, we have for several years been exploring the possibilities of visually guided robots with low-bandwidth vision. Rather than design the robot controllers by hand, we use artificial evolution (in the form of an extended genetic algorithm) to automatically generate the architectures for artificial neural networks which generate effective sensory-motor coordination when controlling mobile robots. Analytic techniques drawn from neuroethology and dynamical systems theory allow us to understand how the evolved robot controllers function, and to predict their behaviour in environments other than those used during the evolutionary process. Initial experiments were performed in simulation, but the techniques have now been successfully transferred to work with a variety of real physical robot platforms. This chapter reviews our past work, concentrating on the analysis of evolved controllers, and gives an overview of our current research. We conclude with a discussion of the application of our evolutionary techniques to problems in biological vision.

We present results from the concurrent evolution of visual sensing morphologies and sensory-motor controller-networks for visually guided robots. In this paper we analyse two (of many) networks which result from using incremental evolution with variable-length genotypes. The two networks come from separate populations, evolved using a common fitness function. The observable behaviours of the two robots are very similar, and close to the optimal behaviour. However, the underlying sensing morphologies and sensory-motor controllers are strikingly different. This is a case of convergent evolution at the behavioural level, coupled with divergent evolution at the morphological level. The action of the evolved networks is described. We discuss the process of analysing evolved artificial networks, a process which bears many similarities to analysing biological nervous systems in the field of neuroethology.

: This paper describes a set of tools developed at our laboratory that provide a reliable set-up for conducting bio-inspired experiments with real robots. We focus on the hardware tools needed to monitor team performances as well as those to achieve collective adaptive behaviours. We propose concrete solutions to some of the main problems in collective robotics. The four main results we derive are: first, the hardware modularity of the miniature robot Khepera [1] allows us to build a flexible set-up; second, the energy autonomy problem is solved in a reliable way for experimenting with real robots during several hours; third, the communication architecture among teammates and/or with a supervisor unit is designed to prevent bandwidth bottlenecks with bigger robot teams; fourth, the use of programmable active pucks (also called "seeds" below) extends the set of possible bio-inspired experiments without increasing the sensorial complexity of the robots. A simple bio-inspired...

This paper discusses two prominent theories of cognitive development and relates them to experiments in social robotics. The main difference between these theories lies in the different views on the relationship between a child and its social environment: a) the child as a solitary thinker (Piaget) and b) the child in society (Vygotsky). We discuss the implications this has on the design of socially intelligent agents, focusing on robotic agents. We argue that the framework proposed by Vygotsky provides a promising research direction in autonomous agents. We give examples of implementations in the area of social robotics which support our theoretical considerations. More specifically, we demonstrate how a teacher-learner setup can be used to teach a robot a proto-language. The same control architecture is also used for a humanoid doll robot which can interact with a human by imitation. Another experiment addresses dynamic coupling of movements between a human and a mobile robot. Here, ...

this memory is implemented by a flag e b for each base, which indicates that this base needs to be examined later. This flag is the state of a competition dynamics similar to (3.9): ø e e b = fi b e b \Gamma jfi b je