Towards an Insect Brain Computational Model - CORDIS... Insect Brain Model As in insects, the...
1
Towards an Insect Brain Computational Model Towards an Insect Brain Computational Model SPARK II Spatial-temporal Patterns for Action-oriented perception in Roving robots www.spark2.diees.unict.it Insect Brain Model Insect Brain Model As in insects, the proposed perceptual architecture is organized in various control levels consisting of functional blocks, acting either at the same level, as competitors, or at distinct hierarchical levels showing the capability to learn more complex, experience-based behaviors. The control architecture consists of series of parallel sensory-motor pathways (i.e. basic behaviours) that are triggered and controlled by specific sensory events in a reflexive way, giving the knowledge baseline to the system. Going up in the hierarchical scheme, two relevant centers of the insect brain are considered: the Mushroom Bodies (MBs) and the Central Complex (CX). Both MBs and CX are not yet well understood from a biological/neurogenetic point of view. However interesting studies underlined how deeply these structures are involved in perceptual processes. In particular MBs are mainly devoted to the enhancement of causal relations arising among the basic behaviours, by exploiting the temporal correlation between sensory events; information storage and retrieval in the case of the olfaction sense; resolving contradictory cues through the visual sense by imposing continuation or adaptive termination of ongoing behaviour. CX is instead responsible for integration and elaboration of visual information, storing and retrieving information on objects and their position in space, controlling the step length in order to approach or avoid such objects; motor control, landmark orientation and navigation, orientation storage and others. Some of this functionalities have been already developed creating a correlation-based anticipation layer. Project Coordinator DIEES University of Catania Prof. Paolo Arena ([email protected]) Ph.D Luca Patanè([email protected]) Tribot Tribot Tribot Tribot II II Sensory Sensory-motor motor control control architecture architecture Spiking Spiking network network for for basic basic behaviours behaviours learning learning At a higher level of the architecture, we introduce a Decision layer that elaborates all the preprocessed sensory information in order to define the final behavior. Memory is of course distributed in the whole architecture but a specific block has been also considered. This block develops a contextual layer where sequences of successful emerged behaviours are memorized to be retrieved when needed. In such a way, as it happens in insects, the basic behaviors, which are often life-saving sensory-motor pathways, are progressively enriched with emergent capabilities which incrementally increase the animal skills. Project Description Project Description The aim of the SPARK II Project is to develop an insect brain inspired computational model as a new architecture for action-oriented perception, inspired by the basic principles of information processing in living systems and exploring the concept of "self-organization". It will take the advantage of the new insights offered by experimental neurobiology regarding the structure and function of relevant centres in the insect brain, devoted to action-oriented perception. These insights will be enriched by the addition of nonlinear spatial temporal dynamical systems able to show emerging patterns used as “perceptual states”. Following a background phase, performed in the former FP6 funded project SPARK, where relevant centres in insects and also different types of spatial-temporal dynamics were investigated, the challenging idea within SPARK II is to introduce a new computational infrastructure mimicking an insect brain architecture. This architecture will be assessed and applied to different robotic structures, in order to prove its generality. The architecture is envisaged to be hierarchical, based on parallel sensory-motor pathways, implementing reflex-driven basic behaviours, enriched with higher and more complex structures, where a mix of bio-inspired artificial neuropils (e.g. for attention-like processes, a short-term memory for planned paths, a memory for dangerous and for rewarding objects) and of more physics-inspired nonlinear lattices, able to generate complex dynamics, work concurrently to generate behaviours at the output motor layer. The architecture will exploit a number of different sensors, processing signals distributed in space and time and also showing nonlinear dynamics. Perceptual processes are conceived as emerging pattern flows (result of a nonlinear spatial-temporal dynamics). Pattern meaning (concept generation) will be incrementally built upon information derived from sensors. It will influence the particular associated motor behaviour with the concurrent dynamics generated by models of the relevant perception centres in insects. The investigation will be focussed on the following two main points: 1) theoretical insights in neurodynamics, both from the point of view of neural networks organised in lattices, and from the biological point of view, with attention to relevant neural areas in insects, like the Mushroom Bodies and the Central Complex, devoted to multimodal interaction and perception; 2) application of the action-oriented perception models to different robotics architectures, like manipulators, mobile and bio-inspired robots. This will show the generality of the proposed approach, suitable for application, with a minor effort, to very different artefacts, so addressing the perceptual process as based on general rules. The ultimate concept that will be proven is that merging complex dynamics and biological inspection in insect brain leads to the emergence of a powerful general system: a new insect brain computational model for perception. Start time : February 2008 End-time : January 2011 Software/Hardware Software/Hardware framework for framework for Cognitive Systems Cognitive Systems Hybrid Robot Hybrid Robot
Transcript of Towards an Insect Brain Computational Model - CORDIS... Insect Brain Model As in insects, the...
Towards an Insect Brain Computational ModelTowards an Insect Brain Computational ModelSPARK II Spatial-temporal Patterns for Action-oriented perception in Roving robots
www.spark2.diees.unict.it
Insect Brain ModelInsect Brain Model
As in insects, the proposed perceptual architecture is organized in various control levels consisting of functional blocks, acting either at the same level, as competitors, or at distinct hierarchical levels showing the capability to learn more complex, experience-based behaviors. The control architecture consists of series of parallel sensory-motor pathways (i.e. basic behaviours) that are triggered and controlled by specific sensory events in a reflexive way, giving the knowledge baseline to the system. Going up in the hierarchical scheme, two relevant centers of the insect brain are considered: the Mushroom Bodies (MBs) and the Central Complex (CX). Both MBs and CX are not yet well understood from a biological/neurogenetic point of view. However interesting studies underlined how deeply these structures are involved in perceptual processes. In particular MBs are mainly devoted to the enhancement of causal relations arising among the basic behaviours, by exploiting the temporal correlation between sensory events;information storage and retrieval in the case of the olfaction sense; resolving contradictory cues through the visual sense by imposing continuation or adaptive termination of ongoing behaviour. CX is instead responsible for integration and elaboration of visual information, storing and retrieving information on objects and their position in space, controlling the step length in order to approach or avoid such objects; motor control, landmark orientation and navigation, orientation storage and others. Some of this functionalities have been already developed creating a correlation-based anticipation layer.
Project CoordinatorDIEES University of Catania
Prof. Paolo Arena ([email protected])Ph.D Luca Patanè([email protected])
TribotTribot
TribotTribot IIII
SensorySensory--motormotor controlcontrolarchitecturearchitecture SpikingSpiking network network forfor basicbasic
behavioursbehaviours learninglearning
At a higher level of the architecture, we introduce a Decision layer that elaborates all the preprocessed sensory information in order to define the final behavior. Memory is of course distributed in the whole architecture but a specific block has been also considered. This block develops a contextual layer where sequences of successful emerged behaviours are memorized to be retrieved when needed. In such a way, as it happens in insects, the basic behaviors, which are often life-saving sensory-motor pathways, are progressively enriched with emergent capabilities which incrementally increase the animal skills.
Project DescriptionProject DescriptionThe aim of the SPARK II Project is to develop an insect brain inspired computational model as a new architecture for action-oriented perception, inspired by the basic principles of information processing in living systems and exploring the concept of "self-organization". It will take the advantage of the new insights offered by experimental neurobiology regarding the structure and function of relevant centres in the insect brain, devoted to action-oriented perception. These insights will be enriched by the addition of nonlinear spatial temporal dynamical systems able to show emerging patterns used as “perceptual states”. Following a background phase, performed in the former FP6 funded project SPARK, where relevant centres in insects and also different types of spatial-temporal dynamics were investigated, the challenging idea within SPARK II is to introduce a new computational infrastructure mimicking an insect brain architecture. This architecture will be assessed and applied to different robotic structures, in order to prove its generality. The architecture is envisaged to be hierarchical, based on parallel sensory-motor pathways, implementing reflex-driven basic behaviours, enriched with higher and more complex structures, where a mix of bio-inspired artificial neuropils (e.g. for attention-like processes, a short-term memory for planned paths, a memory for dangerous and for rewarding objects) and of more physics-inspired nonlinear lattices, able to generate complex dynamics, work concurrently to generate behaviours at the output motor layer. The architecture will exploit a number of different sensors, processing signals distributed in space and time and also showing nonlinear dynamics. Perceptual processes are conceived as emerging pattern flows (result of a nonlinear spatial-temporal dynamics). Pattern meaning (concept generation) will be incrementally built upon information derived from sensors. It will influence the particular associated motor behaviour with the concurrent dynamics generated by models of the relevant perception centres in insects.
The investigation will be focussed on the following two main points:
1) theoretical insights in neurodynamics, both from the point of view of neural networks organised in lattices, and from the biological point of view, with attention to relevant neural areas in insects, like the Mushroom Bodies and the Central Complex, devoted to multimodal interaction and perception;
2) application of the action-oriented perception models to different robotics architectures, like manipulators, mobile and bio-inspired robots. This will show the generality of the proposed approach, suitable for application, with a minor effort, to very different artefacts, so addressing the perceptual process as based on general rules.
The ultimate concept that will be proven is that merging complex dynamics and biological inspection in insect brain leads to the emergence of a powerful general system: a new insect brain computational model for perception.
Start time : February 2008 End-time : January 2011
Software/Hardware Software/Hardware framework for framework for
Cognitive SystemsCognitive SystemsHybrid RobotHybrid Robot
