[show abstract][hide abstract] ABSTRACT: This study focuses on biomimetic sensory motor control of a robotic arm. We have developed a command circuit that was mathematically deduced from physical and mathematical constraints describing the function of cerebellar pathways. The control circuit contains an internal predictive model of the direct biomechanical function of the limb placed in a closed loop, so that the circuit computes an approximate inverse function. The structure of the model resembles the anatomic connectivity of the cerebellar pathways. In this paper, we present an application of this model to the control of a 2-link robotic arm actuated by four single-joint McKibben muscles and report the results obtained by simulation and real-time learning of 2 degrees of freedom pointing movements.
IEEE International Conference on Robotics and Automation, ICRA 2011, Shanghai, China, 9-13 May 2011; 01/2011
[show abstract][hide abstract] ABSTRACT: Several authors suggested that gravitational forces are centrally represented in the brain for planning, control and sensorimotor predictions of movements. Furthermore, some studies proposed that the cerebellum computes the inverse dynamics (internal inverse model) whereas others suggested that it computes sensorimotor predictions (internal forward model).
This study proposes a model of cerebellar pathways deduced from both biological and physical constraints. The model learns the dynamic inverse computation of the effect of gravitational torques from its sensorimotor predictions without calculating an explicit inverse computation. By using supervised learning, this model learns to control an anthropomorphic robot arm actuated by two antagonists McKibben artificial muscles. This was achieved by using internal parallel feedback loops containing neural networks which anticipate the sensorimotor consequences of the neural commands. The artificial neural networks architecture was similar to the large-scale connectivity of the cerebellar cortex. Movements in the sagittal plane were performed during three sessions combining different initial positions, amplitudes and directions of movements to vary the effects of the gravitational torques applied to the robotic arm. The results show that this model acquired an internal representation of the gravitational effects during vertical arm pointing movements.
This is consistent with the proposal that the cerebellar cortex contains an internal representation of gravitational torques which is encoded through a learning process. Furthermore, this model suggests that the cerebellum performs the inverse dynamics computation based on sensorimotor predictions. This highlights the importance of sensorimotor predictions of gravitational torques acting on upper limb movements performed in the gravitational field.
PLoS ONE 02/2009; 4(4):e5176. · 3.73 Impact Factor
[show abstract][hide abstract] ABSTRACT: To address the problem of how the cerebellum processes the premotor orders that control fast movements of the forearm, a model of the cerebellar control is proposed: a cybernetic circuit composed of a model of the cerebellar premotor pathways driving a biomechanical model of the human forearm. Experiments consist of recording electromyographic (EMG) activities and cinematic variables of the human forearm during fast, single joint, point-to-point movements performed in horizontal and vertical directions with and without mass. The biomechanical model of the forearm is first validated by comparing actual movements and movements simulated by using, as inputs to this model, the synthesized EMG signals and of real EMG activities recorded during the experiments. Then the entire control model is validated by comparing actual movements to the desired ones simulated by the model of the cerebellar pathways whose inputs are velocity signals with Gaussian time-courses. The results show that approximate inverse functions can be computed by means of inner models of direct functions placed in feedback loops, and suggest that the orientation of any member segment with respect to gravity is computed as a cinematic variable in the Central Nervous System (CNS).
Journal of Integrative Neuroscience 01/2009; 7(4):481-500. · 1.15 Impact Factor
[show abstract][hide abstract] ABSTRACT: An important question in the literature focusing on motor control is to determine which laws drive biological limb movements. This question has prompted numerous investigations analyzing arm movements in both humans and monkeys. Many theories assume that among all possible movements the one actually performed satisfies an optimality criterion. In the framework of optimal control theory, a first approach is to choose a cost function and test whether the proposed model fits with experimental data. A second approach (generally considered as the more difficult) is to infer the cost function from behavioral data. The cost proposed here includes a term called the absolute work of forces, reflecting the mechanical energy expenditure. Contrary to most investigations studying optimality principles of arm movements, this model has the particularity of using a cost function that is not smooth. First, a mathematical theory related to both direct and inverse optimal control approaches is presented. The first theoretical result is the Inactivation Principle, according to which minimizing a term similar to the absolute work implies simultaneous inactivation of agonistic and antagonistic muscles acting on a single joint, near the time of peak velocity. The second theoretical result is that, conversely, the presence of non-smoothness in the cost function is a necessary condition for the existence of such inactivation. Second, during an experimental study, participants were asked to perform fast vertical arm movements with one, two, and three degrees of freedom. Observed trajectories, velocity profiles, and final postures were accurately simulated by the model. In accordance, electromyographic signals showed brief simultaneous inactivation of opposing muscles during movements. Thus, assuming that human movements are optimal with respect to a certain integral cost, the minimization of an absolute-work-like cost is supported by experimental observations. Such types of optimality criteria may be applied to a large range of biological movements.
[show abstract][hide abstract] ABSTRACT: This article addresses the relationships between motion sickness (MS) and three-dimensional (3D) ocular responses during otolith stimulation. A group of 19 healthy subjects was tested for motion sickness during a 16 min otolith stimulation induced by off-vertical axis rotation (OVAR) (constant velocity 60 degrees /s, frequency 0.16 Hz). For each subject, the MS induced during the session was quantified, and based on this quantification, the subjects were divided into two groups of less susceptible (MS-), and more susceptible (MS+) subjects. The angular eye velocity induced by the otolith stimulation was analyzed in order to identify a possible correlation between susceptibility to MS and 3D eye velocity. The main results show that: (1) MS significantly correlates in a multiple regression with several components of the horizontal vestibular eye movements i.e. positively with the velocity modulation (P<0.01) and bias (P<0.05) of the otolith ocular reflex and negatively with the time constant of the vestibulo-ocular reflex (P<0.01) and (2) the length of the resultant 3D eye velocity vector is significantly larger in the MS+ as compared with the MS- group. Based on these results we suggest that the CNS, including the velocity storage mechanism, reconstructs an eye velocity vector modulated by head position whose length might predict MS occurrence during OVAR.
[show abstract][hide abstract] ABSTRACT: This article has been withdrawn consistent with Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy). The Publisher apologizes for any inconvenience this may cause.
[show abstract][hide abstract] ABSTRACT: RESUME
Pour adresser le problème de façon dont le cervelet traite les commandes prémotrices des mouvements rapides de l'avant-bras nous proposons un modèle du contrôle cérébelleux : un circuit cybernétique qui se compose de modèle des voies cérébelleuses conduisant un modèle biomécanique de l'avant-bras humain. Les expériences se composent de l'enregistrement des activités EMG et des variables cinématiques des mouvements rapides de pointage de 1 degré de liberté effectués dans des plans horizontaux et verticaux. Le modèle biomécanique de l'avant-bras est d'abord validé en comparant les mouvements réels et les mouvements simulés par utilisation, comme entrées à ce modèle, des signaux synthétisés d'EMG et des activités réelles d'EMG enregistrées pendant les expériences. Par la suite le modèle entier de commande est validé en comparant les mouvements désirés et les mouvements accomplis par le modèle dont les entrées sont des signaux de vitesse du mouvement désiré. Les résultats prouvent que des fonctions inverses approximatives peuvent être calculées au moyen de modèles intérieurs des fonctions directes placées dans des boucles de rétroaction, et suggèrent que l'orientation de n'importe quel segment de membre par rapport à la pesanteur soit calculée comme variable cinématique dans le système nerveux central.
MOTS CLES: Neurosciences Computationnelles, Contrôle sensori moteur, Cervelet, Mouvements de pointage, EMG,
[show abstract][hide abstract] ABSTRACT: The command and control of limb movements by the cerebellar and reflex pathways are modeled by means of a circuit whose structure is deduced from functional constraints. One constraint is that fast limb movements must be accurate although they cannot be continuously controlled in closed loop by use of sensory signals. Thus, the pathways which process the motor orders must contain approximate inverse functions of the bio-mechanical functions of the limb and of the muscles. This can be achieved by means of parallel feedback loops, whose pattern turns out to be comparable to the anatomy of the cerebellar pathways. They contain neural networks able to anticipate the motor consequences of the motor orders, modeled by artificial neural networks whose connectivity is similar to that of the cerebellar cortex. These networks learn the direct biomechanical functions of the limbs and muscles by means of a supervised learning process. Teaching signals calculated from motor errors are sent to the learning sites, as, in the cerebellum, complex spikes issued from the inferior olive are conveyed to the Purkinje cells by climbing fibers. Learning rules are deduced by a differential calculation, as classical gradient rules, and they account for the long term depression which takes place in the dendritic arborizations of the Purkinje cells. Another constraint is that reflexes must not impede voluntary movements while remaining at any instant ready to oppose perturbations. Therefore, efferent copies of the motor orders are sent to the interneurones of the reflexes, where they cancel the sensory-motor consequences of the voluntary movements. After learning, the model is able to drive accurately, both in velocity and position, angular movements of a rod actuated by two pneumatic McKibben muscles. Reflexes comparable to the myotatic and tendinous reflexes, and stabilizing reactions comparable to the cerebellar sensory-motor reactions, reduce efficiently the effects of perturbing torques. These results allow to link the behavioral concepts of the equilibrium-point "lambda model" [J Motor Behav 18 (1986) 17] with anatomical and physiological features: gains of reflexes and sensori-motor reactions set the slope of the "invariant characteristic," and efferent copies set the "threshold of the stretch reflex." Thus, mathematical and physical laws account for the raison d'etre of the inhibitory nature of Purkinje cells and for the conspicuous anatomical pattern of the cerebellar pathways. These properties of these pathways allow to perform approximate inverse calculations after learning of direct functions, and insure also the coordination of voluntary and reflex motor orders.
[show abstract][hide abstract] ABSTRACT: A control circuit is proposed to model the command of saccadic eye movements. Its wiring is deduced from a mathematical constraint, i.e. the necessity, for motor orders processing, to compute an approximate inverse function of the bio-mechanical function of the moving plant, here the bio-mechanics of the eye. This wiring is comparable to the anatomy of the cerebellar pathways. A predicting element, necessary for inversion and thus for movement accuracy, is modeled by an artificial neural network whose structure, deduced from physical constraints expressing the mechanics of the eye, is similar to the cell connectivity of the cerebellar cortex. Its functioning is set by supervised reinforcement learning, according to learning rules aimed at reducing the errors of pointing, and deduced from a differential calculation. After each movement, a teaching signal encoding the pointing error is distributed to various learning sites, as is, in the cerebellum, the signal issued from the inferior olive and conveyed to various cell types by the climbing fibers. Results of simulations lead to predict the existence of a learning site in the glomeruli. After learning, the model is able to accurately simulate saccadic eye movements. It accounts for the function of the cerebellar pathways and for the final integrator of the oculomotor system. The novelty of this model of movement control is that its structure is entirely deduced from mathematical and physical constraints, and is consistent with general anatomy, cell connectivity and functioning of the cerebellar pathways. Even the learning rules can be deduced from calculation, and they reproduce long term depression, the learning process which takes place in the dendritic arborization of the Purkinje cells. This approach, based on the laws of mathematics and physics, appears thus as an efficient way of understanding signal processing in the motor system.
[show abstract][hide abstract] ABSTRACT: This article describes an expanded version of a previously proposed motor control scheme, based on rules for combining sensory and motor signals within the central nervous system. Classical control elements of the previous cybernetic circuit were replaced by artificial neural network modules having an architecture based on the connectivity of the cerebellar cortex, and whose functioning is regulated by reinforcement learning. The resulting model was then applied to the motion control of a mechanical, single-joint robot arm actuated by two McKibben artificial muscles. Various biologically plausible learning schemes were studied using both simulations and experiments. After learning, the model was able to accurately pilot the movements of the robot arm, both in velocity and position.
[show abstract][hide abstract] ABSTRACT: The sensory weighting model is a general model of sensory integration that consists of three processing layers. First, each sensor provides the central nervous system (CNS) with information regarding a specific physical variable. Due to sensor dynamics, this measure is only reliable for the frequency range over which the sensor is accurate. Therefore, we hypothesize that the CNS improves on the reliability of the individual sensor outside this frequency range by using information from other sensors, a process referred to as "frequency completion." Frequency completion uses internal models of sensory dynamics. This "improved" sensory signal is designated as the "sensory estimate" of the physical variable. Second, before being combined, information with different physical meanings is first transformed into a common representation; sensory estimates are converted to intermediate estimates. This conversion uses internal models of body dynamics and physical relationships. Third, several sensory systems may provide information about the same physical variable (e.g., semicircular canals and vision both measure self-rotation). Therefore, we hypothesize that the "central estimate" of a physical variable is computed as a weighted sum of all available intermediate estimates of this physical variable, a process referred to as "multicue weighted averaging." The resulting central estimate is fed back to the first two layers. The sensory weighting model is applied to three-dimensional (3D) visual-vestibular interactions and their associated eye movements and perceptual responses. The model inputs are 3D angular and translational stimuli. The sensory inputs are the 3D sensory signals coming from the semicircular canals, otolith organs, and the visual system. The angular and translational components of visual movement are assumed to be available as separate stimuli measured by the visual system using retinal slip and image deformation. In addition, both tonic ("regular") and phasic ("irregular") otolithic afferents are implemented. Whereas neither tonic nor phasic otolithic afferents distinguish gravity from linear acceleration, the model uses tonic afferents to estimate gravity and phasic afferents to estimate linear acceleration. The model outputs are the internal estimates of physical motion variables and 3D slow-phase eye movements. The model also includes a smooth pursuit module. The model matches eye responses and perceptual effects measured during various motion paradigms in darkness (e.g., centered and eccentric yaw rotation about an earth-vertical axis, yaw rotation about an earth-horizontal axis) and with visual cues (e.g., stabilized visual stimulation or optokinetic stimulation).
[show abstract][hide abstract] ABSTRACT: In this article we present a biologically inspired motor control scheme based on sensory-motor interaction modalities within the central nervous system, and its application to the control of a single joint limb segment actuated by two pneumatic McKibben muscles. The embedded artificial neural network module's architecture, whose functioning is regulated by reinforcement learning, is similar to the connectivity of cerebellar cortex. Various biologically plausible learning schemes, which enable functional plasticity in the cerebellar cortex, are discussed. The simulation and experimental results are then reported.
Engineering in Medicine and Biology Society, 2001. Proceedings of the 23rd Annual International Conference of the IEEE; 02/2001
[show abstract][hide abstract] ABSTRACT: Since motion sickness (MS) never occurs in individuals who lack functional vestibular apparatus, it has been suggested that MS susceptible individuals have more sensitive vestibular systems than non-susceptible people. However, previous investigations involving only stimulation of the semi-circular canals have been inconclusive. We measured gain and time constant (TC) of horizontal canal-ocular reflex (COR) and magnitude of otolith-ocular reflex (OOR). We found that MS susceptibility was not correlated to COR gain but was negatively correlated to OOR magnitude. Thus, MS susceptible individuals do not have more sensitive vestibular systems. We also found a positive correlation between MS susceptibility and TC. We hypothesize that central vestibular integration (velocity storage mechanism), by increasing low frequency vestibular inputs, would favour MS.
[show abstract][hide abstract] ABSTRACT: Off-vertical axis rotation (OVAR) at constant velocity is a dynamic otolith stimulus that induces horizontal and vertical eye movement responses. To determine the value of this examination as a test for unilateral otolithic hypofunction, we compared the OVAR responses of patients suffering from acute vestibular neuritis (VN) without any sign of otolith affection, with those of patients suffering from acute VN with otolithic signs. The horizontal eye movement bias component shows directional preponderance (DP) significantly higher in patients with otolithic signs than in patients not presenting them. However, as bias DP also reflects the imbalance between right and left horizontal canals activity, this greater bias DP could be explained by the more severe canals impairment-evaluated by caloric test-found in patients with otolithic signs. No significant difference can be shown on horizontal modulation. The DP of vertical modulation is significantly higher in patients presenting otolithic signs than in patients not presenting them: in the case of otolithic signs, the responses are smaller during rotations toward the affected side. Therefore, this variable could be used as an indication of unilateral otolithic hypofunction.
[show abstract][hide abstract] ABSTRACT: Accuracy of movements requires that the central nervous system computes approximate inverse functions of the mechanical functions of limb articulations. In vertebrates, this is known to be achieved within the cerebellar pathways, and also in the cerebral cortex of primates. A cybernetic circuit achieving this computation allows accurate simulation of fast movements of the eye or forearm. It is consistent with anatomy, and with the classical view of the cerebellum as permanently supervised by the inferior olive. The inferior olive detects over-or under-shoots of movements, and the resulting climbing fiber activity corrects ongoing movements, regulates the function of cerebellar cortex and nuclei, and sets the gains of the sensorimotor reactions.
[show abstract][hide abstract] ABSTRACT: Off vertical axis rotation (OVAR) is a stimulus that can be used to assess the otolith-ocular reflex. However, experimental data suggest that isolated unilateral lesion of the lateral semicircular canal (SCC) nerve could modify responses to OVAR. Thus, to determine what nystagmus variables are not affected by SCC dysfunction and might be used as indices of otolithic disease, responses to OVAR were compared in 39 healthy controls and in 19 patients suffering from acute unilateral vestibular neuritis (VN), without any sign of otolith dysfunction. Horizontal and vertical slow phase velocities (SPV) were measured during earth vertical axis rotation (EVAR), and during OVAR at a tilt angle of 9 degrees and rotation velocity of 60 degrees/s. During OVAR, horizontal SPV consists of a sinusoidal modulation superimposed on a sustained bias opposite to the rotation. Vertical SPV consists of a sinusoidal modulation without bias. In patients, the bias shows directional preponderance (DP) toward the healthy side, strongly correlated to EVAR nystagmus DP. It would therefore simply reflect an imbalance, produced by the unilateral peripheral vestibular lesion, between right and left vestibular nuclei activity. On the other hand, vertical and horizontal modulations are not significantly different in patients and controls. Since the cause and the site of VN are not known, we cannot be sure that patients had pure SCC deafferentation. However, as all of them had SCC paresis it is concluded that OVAR modulations are not affected by a strong dysfunction of the pathways issued from the SCCs.
[show abstract][hide abstract] ABSTRACT: Motion sickness (MS) susceptibility of 108 normal subjects was measured during off-vertical axis rotation (OVAR) as a function of angular velocity (60-180 degrees/s). The chair rotated about a longitudinal axis tilted 30 degrees with respect to gravity. For each velocity, we measured the duration of exposure necessary to evoke a moderate malaise, with a limit of 30 min. MS appeared the fastest at a rotation velocity of 105 degrees/s; higher or lower velocities were less provocative. These results are in good agreement with predictions made by Zupan et al. [in ICANN'94, Springer-Verlag, 1995] by means of a MS mathematical model derived from a model of sensory interactions [Droulez and Darlot, in Attention and Performance, Vol. 13, Lawrence Erlbaum, Hillsdale, 1989]. We also found that MS susceptibility during OVAR is positively correlated with susceptibility to other forms of MS. Since OVAR induces sensory messages very different from those induced by other provocative stimulations, this could suggest that the sensitivity of a common final vegetative locus is an important factor of the individual differences in susceptibility to MS.