Sensory prediction as a role for the cerebellum

Tutorial Review of Bio-Inspired Approaches to Robotic Manipulation for Space Debris Salvage

We present a comprehensive tutorial review that explores the application of bio-inspired approaches to robot control systems for grappling and manipulating a wide range of space debris targets. Current robot manipulator control systems exploit limited techniques which can be supplemented by additional bio-inspired methods to provide a robust suite of robot manipulation technologies. In doing so, we review bio-inspired control methods because this will be the key to enabling such capabilities. In particular, force feedback control may be supplemented with predictive forward models and software emulation of viscoelastic preflexive joint behaviour. This models human manipulation capabilities as implemented by the cerebellum and muscles/joints respectively. In effect, we are proposing a three-level control strategy based on biomimetic forward models for predictive estimation, traditional feedback control and biomimetic muscle-like preflexes. We place emphasis on bio-inspired forward modelling suggesting that all roads lead to this solution for robust and adaptive manipulator control. This promises robust and adaptive manipulation for complex tasks in salvaging space debris.

Neural mechanisms for prediction: do insects have forward models?

'Forward models' are increasingly recognized as a crucial explanatory concept in vertebrate motor control. The essential idea is that an important function implemented by nervous systems is prediction of the sensory consequences of action. This is often associated with higher cognitive capabilities; however, many of the purposes forward models are thought to have analogues in insect behaviour, and the concept is closely connected to those of 'efference copy' and 'corollary discharge'. This article considers recent evidence from invertebrates that demonstrates the predictive modulation of sensory processes by motor output, and discusses to what extent this supports the conclusion that insect nervous systems also implement forward models. Several promising directions for further research are outlined.

Outputs of radula mechanoafferent neurons in Aplysia are modulated by motor neurons, interneurons, and sensory neurons

The gain of sensory inputs into the nervous system can be modulated so that the nature and intensity of afferent input is variable. Sometimes the variability is a function of other sensory inputs or of the state of motor systems that generate behavior. A form of sensory modulation was investigated in the Aplysia feeding system at the level of a radula mechanoafferent neuron (B21) that provides chemical synaptic input to a group of motor neurons (B8a/b, B15) that control closure and retraction movements of the radula, a food grasping structure. B21 has been shown to receive both excitatory and inhibitory synaptic inputs from a variety of neuron types. The current study investigated the morphological basis of these heterosynaptic inputs, whether the inputs could serve to modulate the chemical synaptic outputs of B21, and whether the neurons producing the heterosynaptic inputs were periodically active during feeding motor programs that might modulate B21 outputs in a phase-specific manner. Four cell types making monosynaptic connections to B21 were found capable of heterosynaptically modulating the chemical synaptic output of B21 to motor neurons B8a and B15. These included the following: 1) other sensory neurons, e.g. , B22; 2) interneurons, e.g., B19; 3) motor neurons, e.g., B82; and 4) multifunction neurons that have sensory, motor, and interneuronal functions, e.g., B4/5. Each cell type was phasically active in one or more feeding motor programs driven by command-like interneurons, including an egestive motor program driven by CBI-1 and an ingestive motor program driven by CBI-2. Moreover, the phase of activity differed for each of the modulator cells. During the motor programs, shifts in B21 membrane potential were related to the activity patterns of some of the modulator cells. Inhibitory chemical synapses mediated the modulation produced by B4/5, whereas excitatory and/or electrical synapses were involved in the other instances. The data indicate that modulation is due to block of action potential invasion into synaptic release regions or to alterations of transmitter release as a function of the presynaptic membrane potential. The results indicate that just as the motor system of Aplysia can be modulated by intrinsic mechanisms that can enhance its efficiency, the properties of primary sensory cells can be modified by diverse inputs from mediating circuitry. Such modulation could serve to optimize sensory cells for the different roles they might play.

Generative character of perception: a neural architecture for sensorimotor anticipation

The basic idea of our anticipatory approach to perception is to avoid the common separation of perception and generation of behavior and to fuse both aspects into a consistent neural process. Our approach tries to explain the phenomenon of perception, in particular, of perception at the level of sensorimotor intelligence, from a behavior-oriented point of view. Perception is assumed to be a generative process of anticipating the course of events resulting from alternative sequences of hypothetically executed actions. By means of this sensorimotor anticipation, it is possible to characterize a visual scenery immediately in categories of behavior, i.e. by a set of actions which describe possible methods of interaction with the objects in the environment. Thus, the competence to perceive a complex situation can be understood as the capability to anticipate the course of events caused by different action sequences. Starting from an abstract description of anticipatory perception and the essential biological evidence for internal simulation, we present two biologically motivated computational models that are able to anticipate and evaluate hypothetically sensorimotor sequences. Both models consider functional aspects of those cortical and subcortical systems that are assumed to be involved in the process of sensory prediction and sensorimotor control. Our first approach, the Model for Anticipation based on Sensory IMagination (MASIM), realizes a sequential search in sensorimotor space using a simple model of lateral cerebellum as sensory predictor. We demonstrate the efficiency of this model approach in the light of visually guided local navigation behaviors of a mobile system. The second approach, the Model for Anticipation based on Cortical Representations (MACOR), is actually still at a conceptual level of realization. We postulate that this model allows a completely parallel search at the neocortical level using assemblies of spiking neurons for grouping, separation, and selection of sensorimotor sequences. Both models are intended as general schemes for anticipation based perception at the level of sensorimotor intelligence.

A neural model of cerebellar learning for arm movement control: cortico-spino-cerebellar dynamics

A neural network model of opponent cerebellar learning for arm movement control is proposed. The model illustrates how a central pattern generator in cortex and basal ganglia, a neuromuscular force controller in spinal cord, and an adaptive cerebellum cooperate to reduce motor variability during multijoint arm movements using mono- and bi-articular muscles. Cerebellar learning modifies velocity commands to produce phasic antagonist bursts at interpositus nucleus cells whose feed-forward action overcomes inherent limitations of spinal feedback control of tracking. Excitation of alpha motoneuron pools, combined with inhibition of their Renshaw cells by the cerebellum, facilitate movement initiation and optimal execution. Transcerebellar pathways are opened by learning through long-term depression (LTD) of parallel fiber-Purkinje cell synapses in response to conjunctive stimulation of parallel fibers and climbing fiber discharges that signal muscle stretch errors. The cerebellar circuitry also learns to control opponent muscles pairs, allowing cocontraction and reciprocal inhibition of muscles. Learning is stable, exhibits load compensation properties, and generalizes better across movement speeds if motoneuron pools obey the size principle. The intermittency of climbing fiber discharges maintains stable learning. Long-term potentiation (LTP) in response to uncorrelated parallel fiber signals enables previously weakened synapses to recover. Loss of climbing fibers, in the presence of LTP, can erode normal opponent signal processing. Simulated lesions of the cerebellar network reproduce symptoms of cerebellar disease, including sluggish movement onsets, poor execution of multijoint plans, and abnormally prolonged endpoint oscillations.