Jacopo Zenzeri

Teaching a humanoid robot to draw `Shapes'

The core cognitive ability to perceive and synthesize ‘shapes’ underlies all our basic interactions with the world, be it shaping one’s fingers to grasp a ball or shaping one’s body while imitating a dance. In this article, we describe our attempts to understand this multifaceted problem by creating a primitive shape perception/synthesis system for the baby humanoid iCub. We specifically deal with the scenario of iCub gradually learning to draw or scribble shapes of gradually increasing complexity, after observing a demonstration by a teacher, by using a series of self evaluations of its performance. Learning to imitate a demonstrated human movement (specifically, visually observed end-effector trajectories of a teacher) can be considered as a special case of the proposed computational machinery. This architecture is based on a loop of transformations that express the embodiment of the mechanism but, at the same time, are characterized by scale invariance and motor equivalence. The following transformations are integrated in the loop: (a) Characterizing in a compact, abstract way the ‘shape’ of a demonstrated trajectory using a finite set of critical points, derived using catastrophe theory: Abstract Visual Program (AVP); (b) Transforming the AVP into a Concrete Motor Goal (CMG) in iCub’s egocentric space; (c) Learning to synthesize a continuous virtual trajectory similar to the demonstrated shape using the discrete set of critical points defined in CMG; (d) Using the virtual trajectory as an attractor for iCub’s internal body model, implemented by the Passive Motion Paradigm which includes a forward and an inverse motor model; (e) Forming an Abstract Motor Program (AMP) by deriving the ‘shape’ of the self generated movement (forward model output) using the same technique employed for creating the AVP; (f) Comparing the AVP and AMP in order to generate an internal performance score and hence closing the learning loop. The resulting computational framework further combines three crucial streams of learning: (1) motor babbling (self exploration), (2) imitative action learning (social interaction) and (3) mental simulation, to give rise to sensorimotor knowledge that is endowed with seamless compositionality, generalization capability and body-effectors/task independence. The robustness of the computational architecture is demonstrated by means of several experimental trials of gradually increasing complexity using a state of the art humanoid platform.

A neural mechanism of synergy formation for whole body reaching

The present study proposes a computational model for the formation of whole body reaching synergy, i.e., coordinated movements of lower and upper limbs, characterized by a focal component (the hand must reach a target) and a postural component (the center of mass must remain inside the support base). The model is based on an extension of the equilibrium point hypothesis that has been called Passive Motion Paradigm (PMP), modified in order to achieve terminal attractor features and allow the integration of multiple constraints. The model is a network with terminal attractor dynamics. By simulating it in various conditions it was possible to show that it exhibits many of the spatio-temporal features found in experimental data. In particular, the motion of the center of mass appears to be synchronized with the motion of the hand and with proportional amplitude. Moreover, the joint rotation patterns can be accounted for by a single functional degree of freedom, as shown by principal component analysis. It is also suggested that recent findings in motor imagery support the idea that the PMP network may represent the motor cognitive part of synergy formation, uncontaminated by the effect of execution.

Integrating proprioceptive assessment with proprioceptive training of stroke patients

Although proprioceptive impairment is likely to affect in a significant manner the capacity of stroke patients to recover functionality of the upper limb, clinical assessment methods in current use are rather crude, with a low level of reliability and a limited capacity to discriminate the relevant features of the deficits. In this paper we describe a new technique based on robot technology, with the goal of providing a reliable, accurate, quantitative evaluation of the position sense in peri-personal space. The proposed technique uses a bimanual, planar robot manipuladum (BdF device), whose handles are grasped by the blindfolded patient: the paretic hand is passively placed in one of 17 positions and the subject is asked to actively match the paretic hand position in space with the other hand. The position sense of the paretic arm and the corresponding deficit of space representation are characterized by means of 7 indicators: 1) positional error; 2) holding force; 3) medio/lateral shift; 4) antero/posterior shift; 5) medio/lateral skew; 6) antero/posterior skew; 7) shrink coefficient. We also show how the same experimental setup can be used for “proprioceptive training”, i.e. for providing robot assistance to the paretic arm that may improve the position sense of the patient. A preliminary, feasibility test has been carried out with one patient and three controls.

Revisiting the body-schema concept in the context of whole-body postural-focal dynamics.

The body-schema concept is revisited in the context of embodied cognition, further developing the theory formulated by Marc Jeannerod that the motor system is part of a simulation network related to action, whose function is not only to shape the motor system for preparing an action (either overt or covert) but also to provide the self with information on the feasibility and the meaning of potential actions. The proposed computational formulation is based on a dynamical system approach, which is linked to an extension of the equilibrium-point hypothesis, called Passive Motor Paradigm: this dynamical system generates goal-oriented, spatio-temporal, sensorimotor patterns, integrating a direct and inverse internal model in a multi-referential framework. The purpose of such computational model is to operate at the same time as a general synergy formation machinery for planning whole-body actions in humanoid robots and/or for predicting coordinated sensory–motor patterns in human movements. In order to illustrate the computational approach, the integration of simultaneous, even partially conflicting tasks will be analyzed in some detail with regard to postural-focal dynamics, which can be defined as the fusion of a focal task, namely reaching a target with the whole-body, and a postural task, namely maintaining overall stability.

Robot-assisted training of the kinesthetic sense: enhancing proprioception after stroke

Proprioception has a crucial role in promoting or hindering motor learning. In particular, an intact position sense strongly correlates with the chances of recovery after stroke. A great majority of neurological patients present both motor dysfunctions and impairments in kinesthesia, but traditional robot and virtual reality training techniques focus either in recovering motor functions or in assessing proprioceptive deficits. An open challenge is to implement effective and reliable tests and training protocols for proprioception that go beyond the mere position sense evaluation and exploit the intrinsic bidirectionality of the kinesthetic sense, which refers to both sense of position and sense of movement. Modulated haptic interaction has a leading role in promoting sensorimotor integration, and it is a natural way to enhance volitional effort. Therefore, we designed a preliminary clinical study to test a new proprioception-based motor training technique for augmenting kinesthetic awareness via haptic feedback. The feedback was provided by a robotic manipulandum and the test involved seven chronic hemiparetic subjects over 3 weeks. The protocol included evaluation sessions that consisted of a psychometric estimate of the subject’s kinesthetic sensation, and training sessions, in which the subject executed planar reaching movements in the absence of vision and under a minimally assistive haptic guidance made by sequences of graded force pulses. The bidirectional haptic interaction between the subject and the robot was optimally adapted to each participant in order to achieve a uniform task difficulty over the workspace. All the subjects consistently improved in the perceptual scores as a consequence of training. Moreover, they could minimize the level of haptic guidance in time. Results suggest that the proposed method is effective in enhancing kinesthetic acuity, but the level of impairment may affect the ability of subjects to retain their improvement in time.

Strategy Switching in the Stabilization of Unstable Dynamics

In order to understand mechanisms of strategy switching in the stabilization of unstable dynamics, this work investigates how human subjects learn to become skilled users of an underactuated bimanual tool in an unstable environment. The tool, which consists of a mass and two hand-held non-linear springs, is affected by a saddle-like force-field. The non-linearity of the springs allows the users to determine size and orientation of the tool stiffness ellipse, by using different patterns of bimanual coordination: minimal stiffness occurs when the two spring terminals are aligned and stiffness size grows by stretching them apart. Tool parameters were set such that minimal stiffness is insufficient to provide stable equilibrium whereas asymptotic stability can be achieved with sufficient stretching, although at the expense of greater effort. As a consequence, tool users have two possible strategies for stabilizing the mass in different regions of the workspace: 1) high stiffness feedforward strategy, aiming at asymptotic stability and 2) low stiffness positional feedback strategy aiming at bounded stability. The tool was simulated by a bimanual haptic robot with direct torque control of the motors. In a previous study we analyzed the behavior of naïve users and we found that they spontaneously clustered into two groups of approximately equal size. In this study we trained subjects to become expert users of both strategies in a discrete reaching task. Then we tested generalization capabilities and mechanism of strategy-switching by means of stabilization tasks which consist of tracking moving targets in the workspace. The uniqueness of the experimental setup is that it addresses the general problem of strategy-switching in an unstable environment, suggesting that complex behaviors cannot be explained in terms of a global optimization criterion but rather require the ability to switch between different sub-optimal mechanisms.

Expert strategy switching in the control of a bimanual manipulandum with an unstable task

The goal of this study is to better understand how the central nervous system switches between alternative stabilization strategies when presented with an unstable task. A haptic, bimanual manipulandum has been used to emulate an unstable task, which requires subjects to stabilize a virtual mass under the action of a saddle force field with two nonlinear springs, whose stiffness increases with the amount of stretch. Subjects learn to position the mass at various target points by adjusting the rest length, and thus the stiffness of the two springs. From a previous study we know that subjects can stabilize the mass by either 1) applying large forces to stretch the springs and increase the mechanical stiffness of the system beyond a critical level or by 2) applying small force impulses that intermittently adjust the position of the mass. In this study we report the performance of a subject who was trained extensively to use one strategy or the other in order to characterize the mechanism of target switching, from the high-stiffness to the low-stiffness regime and back.

A new method for evaluating kinesthetic acuity during haptic interaction

SUMMARY Although proprioceptive impairment is likely to affect in a significant manner the capacity of stroke patients to recover functionality of upper limb, clinical assessment methods currently in use are rather crude, with a low level of reliability and a limited capacity to discriminate the relevant features of this severe deficit. In the present paper, we describe a new technique based on robot technology, with the goal of providing a reliable, accurate, and quantitative evaluation of kinesthetic acuity, which can be integrated in robot therapy. The proposed technique, based on a pulsed assistance paradigm, has been evaluated on a group of healthy subjects.

Pulsed assistance: A new paradigm of robot training

In this preliminary study we compare continuous with pulsed robot assistance in five chronic stroke survivors with a mild degree of spasticity, with the aim of promoting volitional effort and reducing assistance during a reaching task. The protocol consists of one familiarization session and a single training session during which a manipulandum provides subjects with pulsed or continuous assistance in random order. The basic level of assistive force is calibrated for each subject and is the same for both modalities; however, the average force during continuous assistance is about twice the average force in pulsed assistance. In spite of this, the results show that pulsed assistance allows subjects to reach similar performance levels as compared to continuous assistance after a single training session. Moreover, we introduce a novel kinematic-based measure to assess voluntary participation of subjects during the rehabilitation task, which is only applicable with pulsed assistance.

Skill Learning and Skill Transfer Mediated by Cooperative Haptic Interaction

It is known that physical coupling between two subjects may be advantageous in joint tasks. However, little is known about how two people mutually exchange information to exploit the coupling. Therefore, we adopted a reversed, novel perspective to the standard one that focuses on the ability of physically coupled subjects to adapt to cooperative contexts that require negotiating a common plan: we investigated how training in pairs on a novel task affects the development of motor skills of each of the interacting partners. The task involved reaching movements in an unstable dynamic environment using a bilateral non-linear elastic tool that could be used bimanually or dyadically. The main result is that training with an expert leads to the greatest performance in the joint task. However, the performance in the individual test is strongly affected by the initial skill level of the partner. Moreover, practicing with a peer rather than an expert appears to be more advantageous for a naive; and motor skills can be transferred to a bimanual context, after training with an expert, only if the non-expert subject had prior experience of the dynamics of the novel task.