Luca Patané

Learning Anticipation via Spiking Networks: Application to Navigation Control

In this paper, we introduce a network of spiking neurons devoted to navigation control. Three different examples, dealing with stimuli of increasing complexity, are investigated. In the first one, obstacle avoidance in a simulated robot is achieved through a network of spiking neurons. In the second example, a second layer is designed aiming to provide the robot with a target approaching system, making it able to move towards visual targets. Finally, a network of spiking neurons for navigation based on visual cues is introduced. In all cases, the robot was assumed to rely on some a priori known responses to low-level sensors (i.e., to contact sensors in the case of obstacles, to proximity target sensors in the case of visual targets, or to the visual target for navigation with visual cues). Based on their knowledge, the robot has to learn the response to high-level stimuli (i.e., range finder sensors or visual input). The biologically plausible paradigm of spike-timing-dependent plasticity (STDP) is included in the network to make the system able to learn high-level responses that guide navigation through a simple unstructured environment. The learning procedure is based on classical conditioning.

A CNN-based chip for robot locomotion control

In this paper, the paradigm of emergent computation is applied to locomotion control in legged robots: the locomotion gait is the result of self-organization of a network of locally coupled nonlinear oscillators. This means to adopt the biological paradigm of central pattern generator (CPG), implemented by using cellular neural networks (CNNs). The whole control strategy is hybrid in the sense that the gait generation is accomplished by a fully analog CNN, while a simple logic unit modulates the behavior of the CNN-based CPG, so that the strategy is suitable to eventually include sensory feedback. The design of a VLSI chip implementing the CNN-based CPG and some experimental results on the chip are presented. The chip is designed using a switched-capacitor technique, fundamental to obtain in a simple and direct way some key features of the hybrid control discussed. The experimental results confirm the suitability of the approach.

Reactive navigation through multiscroll systems: from theory to real-time implementation

Abstract In this paper a new reactive layer for multi-sensory integration applied to robot navigation is proposed. The new robot navigation technique exploits the use of a chaotic system able to be controlled in real-time towards less complex orbits, like periodic orbits or equilibrium points, considered as perceptive orbits. These are subject to real-time modifications on the basis of environment changes acquired through a distributed sensory system. The strategy is inspired to the olfactory bulb neural activity observed in rabbits subject to external stimuli. The mathematical details of the approach are given including simulation results in a virtual environment. Furthermore the proposed strategy has been tested on an experimental environment consisting of an FPGA-based hardware driving an autonomous roving robot. The obtained results demonstrate the capability to perform a real-time navigation control.

Sensory feedback in CNN-based central pattern generators.

Central Pattern Generators (CPGs) are a suitable paradigm to solve the problem of locomotion control in walking robots. CPGs are able to generate feed-forward signals to achieve a proper coordination among the robot legs. In literature they are often modelled as networks of coupled nonlinear systems. However the topic of feedback in these systems is rarely addressed. On the other hand feedback is essential for locomotion. In this paper the CPG for a hexapod robot is implemented through Cellular Neural Networks (CNNs). Feedback is included in the CPG controller by exploiting the dynamic properties of the CPG motor-neurons, such as synchronization issue and local bifurcations. These universal paradigms provide the essential issues to include sensory feedback in CPG architectures based on coupled nonlinear systems. Experiments on a dynamic model of a hexapod robot are presented to validate the approach introduced.

2012 Special Issue: Learning expectation in insects: A recurrent spiking neural model for spatio-temporal representation

Insects are becoming a reference point in Neuroscience for the study of biological aspects at the basis of cognitive processes. These animals have much simpler brains with respect to higher animals, showing, at the same time, impressive capability to adaptively react and take decisions in front of complex environmental situations. In this paper we propose a neural model inspired by the insect olfactory system, with particular attention to the fruit fly Drosophila melanogaster. This architecture is a multilayer spiking network, where each layer is inspired by the structures of the insect brain mainly involved in olfactory information processing, namely the Mushroom Bodies, the Lateral Horns and the Antennal Lobes. In the Antennal Lobes layer olfactory signals lead to a competition among sets of neurons, resulting in a pattern which is projected to the Mushroom Bodies layer. Here a competitive reaction-diffusion process leads to a spontaneous emerging of clusters. The Lateral Horns have been modeled as a delayed input-triggered resetting system. Using plastic recurrent connections, with the addition of simple learning mechanisms, the structure is able to realize a top-down modulation at the input level. This leads to the emergence of an attentional loop as well as to the arousal of basic expectation behaviors in case of subsequently presented stimuli. Simulation results and analysis on the biological plausibility of the architecture are provided and the role of noise in the network is reported.

An autonomous mini-hexapod robot controlled through a CNN-based CPG VLSI chip

This paper deals with the application of a VLSI chip for locomotion control of a mini-hexapod robot. The chip that implements a CPG through a CNN-based structure, has been already used in different bio-inspired structures. The aim of this work is to show the realization of a completely autonomous mini-legged robot by means of the VLSI device. Moreover the chip permits the integration of exteroceptive sensors that can be used to close the feedback loop with the environment implementing simple navigation control strategies. Experimental results obtained in different environmental conditions are reported and discussed

Climbing obstacle in bio-robots via CNN and adaptive attitude control

In this paper, we introduce a novel control system architecture for hexapod robots. Our aim is to guarantee efficient horizontal walking and obstacle climbing via suitable postural adjustments. The control scheme takes its inspiration from recent neurobiological and kinematic observations of cockroaches walking on a treadmill and climbing over barriers. Based on a hierarchical and modular approach, the control architecture is divided into two levels. In the low level two parts working in parallel are present: rhythmic movements leading to gaits are performed by a cellular neural network (CNN) playing the role of an artificial central pattern generator (CPG), while a parallel PD attitude control system modulates (with adding terms) the CNN-CPG signals to achieve postural adjustments. The higher level, in turn, adds plasticity to the whole system; it is based on motor maps and maps sensory information in suitable attitude references for the low level PD attitude control. Tests performed with a dynamic model of hexapod have shown that after a training period the high level is able to enhance walking and climbing capabilities. Copyright

A cricket-inspired Neural Network For FeedForward Compensation and Multisensory Integration

A nonlinear feedforward compensator was designed as part of a bioinspired neural network to model sensorimotor integration and control in crickets. Female crickets perform auditory orientation (phonotaxis) towards the males calling song to find a mate. Crickets also use visual sensing, for example in the optomotor reflex which allows them to maintain a straight trajectory against disturbances. The compensator describe in this paper allows the efficient integration of the phonotaxis and optomotor systems. The design is inspired by the neurophysiological concepts of efferent-copy and corollary discharge, which can be directly intepreted within control theory as feedforward compensation for predictable disturbances. The aim is to predict the reafferent visual stimulus caused by phonotaxis, based on the efferent response, thus filtering out the optical disturbances induced by the phonotactic reflex, while still detecting any external noise. The feedforward compensator design was formulated as an identification problem, drawing data from experiments on a robot performing phonotaxis. The compensator parameters were first derived by trial-and-error, and then optimised using a genetic algorithm. The scheme is implemented in a bioinspired neural network on a robot, and experiments are carried out to compare the behaviour to the cricket.

CPG-MTA implementation for locomotion control

This paper deals with the new realization of the VLSI chip for locomotion control introduced in Arena et al. (2003). With respect to the previous version, the new chip includes the fully functional generation of several locomotion gaits suitable for the control of hexapod robots and other bio-inspired structures. Experimental results and the application of the chip to the control of two bio-inspired robots are shown.

An insect-inspired bionic sensor for tactile localization and material classification with state-dependent modulation

Insects carry a pair of antennae on their head: multimodal sensory organs that serve a wide range of sensory-guided behaviors. During locomotion, antennae are involved in near-range orientation, for example in detecting, localizing, probing, and negotiating obstacles. Here we present a bionic, active tactile sensing system inspired by insect antennae. It comprises an actuated elastic rod equipped with a terminal acceleration sensor. The measurement principle is based on the analysis of damped harmonic oscillations registered upon contact with an object. The dominant frequency of the oscillation is extracted to determine the distance of the contact point along the probe and basal angular encoders allow tactile localization in a polar coordinate system. Finally, the damping behavior of the registered signal is exploited to determine the most likely material. The tactile sensor is tested in four approaches with increasing neural plausibility: first, we show that peak extraction from the Fourier spectrum is sufficient for tactile localization with position errors below 1%. Also, the damping property of the extracted frequency is used for material classification. Second, we show that the Fourier spectrum can be analysed by an Artificial Neural Network (ANN) which can be trained to decode contact distance and to classify contact materials. Thirdly, we show how efficiency can be improved by band-pass filtering the Fourier spectrum by application of non-negative matrix factorization. This reduces the input dimension by 95% while reducing classification performance by 8% only. Finally, we replace the FFT by an array of spiking neurons with gradually differing resonance properties, such that their spike rate is a function of the input frequency. We show that this network can be applied to detect tactile contact events of a wheeled robot, and how detrimental effects of robot velocity on antennal dynamics can be suppressed by state-dependent modulation of the input signals.