Renaud Ronsse

Human–Robot Synchrony: Flexible Assistance Using Adaptive Oscillators

We propose a novel method for movement assistance that is based on adaptive oscillators, i.e., mathematical tools that are capable of extracting the high-level features (amplitude, frequency, and offset) of a periodic signal. Such an oscillator acts like a filter on these features, but keeps its output in phase with respect to the input signal. Using a simple inverse model, we predicted the torque produced by human participants during rhythmic flexion extension of the elbow. Feeding back a fraction of this estimated torque to the participant through an elbow exoskeleton, we were able to prove the assistance efficiency through a marked decrease of the biceps and triceps electromyography. Importantly, since the oscillator adapted to the movement imposed by the user, the method flexibly allowed us to change the movement pattern and was still efficient during the nonstationary epochs. This method holds promise for the development of new robot-assisted rehabilitation protocols because it does not require prespecifying a reference trajectory and does not require complex signal sensing or single-user calibration: the only signal that is measured is the position of the augmented joint. In this paper, we further demonstrate that this assistance was very intuitive for the participants who adapted almost instantaneously.

Oscillator-based assistance of cyclical movements: model-based and model-free approaches.

In this article, we propose a new method for providing assistance during cyclical movements. This method is trajectory-free, in the sense that it provides user assistance irrespective of the performed movement, and requires no other sensing than the assisting robot’s own encoders. The approach is based on adaptive oscillators, i.e., mathematical tools that are capable of learning the high level features (frequency, envelope, etc.) of a periodic input signal. Here we present two experiments that we recently conducted to validate our approach: a simple sinusoidal movement of the elbow, that we designed as a proof-of-concept, and a walking experiment. In both cases, we collected evidence illustrating that our approach indeed assisted healthy subjects during movement execution. Owing to the intrinsic periodicity of daily life movements involving the lower-limbs, we postulate that our approach holds promise for the design of innovative rehabilitation and assistance protocols for the lower-limb, requiring little to no user-specific calibration.

Sensing Pressure Distribution on a Lower-Limb Exoskeleton Physical Human-Machine Interface

A sensory apparatus to monitor pressure distribution on the physical human-robot interface of lower-limb exoskeletons is presented. We propose a distributed measure of the interaction pressure over the whole contact area between the user and the machine as an alternative measurement method of human-robot interaction. To obtain this measure, an array of newly-developed soft silicone pressure sensors is inserted between the limb and the mechanical interface that connects the robot to the user, in direct contact with the wearer’s skin. Compared to state-of-the-art measures, the advantage of this approach is that it allows for a distributed measure of the interaction pressure, which could be useful for the assessment of safety and comfort of human-robot interaction. This paper presents the new sensor and its characterization, and the development of an interaction measurement apparatus, which is applied to a lower-limb rehabilitation robot. The system is calibrated, and an example its use during a prototypical gait training task is presented.

Motor learning-induced changes in functional brain connectivity as revealed by means of graph-theoretical network analysis

Complex bimanual motor learning causes specific changes in activation across brain regions. However, there is little information on how motor learning changes the functional connectivity between these regions, and whether this is influenced by different sensory feedback modalities. We applied graph-theoretical network analysis (GTNA) to examine functional networks based on motor-task-related fMRI activations. Two groups learned a complex 90° out-of-phase bimanual coordination pattern, receiving either visual or auditory feedback. 3T fMRI scanning occurred before (day 0) and after (day 5) training. In both groups, improved motor performance coincided with increased functional network connectivity (increased clustering coefficients, higher number of network connections and increased connection strength, and shorter communication distances). Day×feedback interactions were absent but, when examining network metrics across all examined brain regions, the visual group had a marginally better connectivity, higher connection strength, and more direct communication pathways. Removal of feedback had no acute effect on the functional connectivity of the trained networks. Hub analyses showed an importance of specific brain regions not apparent in the standard fMRI analyses. These findings indicate that GTNA can make unique contributions to the examination of functional brain connectivity in motor learning.

Sensorless stabilization of bounce juggling

The paper studies the properties of a sinusoidally vibrating wedge billiard as a model for 2-D bounce juggling. It is shown that some periodic orbits that are unstable in the elastic fixed wedge become exponentially stable in the nonelastic vibrating wedge. These orbits are linked with certain classical juggling patterns, providing an interesting benchmark for the study of the frequency-locking properties in human rhythmic tasks. Experimental results on sensorless stabilization of juggling patterns are described.

A computational model for rhythmic and discrete movements in uni-and bimanual coordination

Current research on discrete and rhythmic movements differs in both experimental procedures and theory, despite the ubiquitous overlap between discrete and rhythmic components in everyday behaviors. Models of rhythmic movements usually use oscillatory systems mimicking central pattern generators (CPGs). In contrast, models of discrete movements often employ optimization principles, thereby reflecting the higher-level cortical resources involved in the generation of such movements. This letter proposes a unified model for the generation of both rhythmic and discrete movements. We show that a physiologically motivated model of a CPG can not only generate simple rhythmic movements with only a small set of parameters, but can also produce discrete movements if the CPG is fed with an exponentially decaying phasic input. We further show that a particular coupling between two of these units can reproduce main findings on in-phase and antiphase stability. Finally, we propose an integrated model of combined rhythmic and discrete movements for the two hands. These movement classes are sequentially addressed in this letter with increasing model complexity. The model variations are discussed in relation to the degree of recruitment of the higher-level cortical resources, necessary for such movements.

Adaptive oscillators with human-in-the-loop: Proof of concept for assistance and rehabilitation

Most recent findings in robot-assisted therapy suggest that the therapy is more successful if the patient actively participates to the movement (“assistance-as-needed”). In the present contribution, we propose a novel approach for designing highly flexible protocols based on this concept. This approach uses adaptive oscillators: a mathematical primitive having the capacity to learn the high-level features of a quasi-sinusoidal signal (amplitude, frequency, offset). Using a simple inverse model, we demonstrate that this method permits to synchronize with the torque produced by the user, such that the effort associated with the movement production is shared between the user and the assistance device, without specifying any arbitrary reference trajectory. Simulation results also establish the method relevance for helping patients with movement disorders. Since our method is specifically designed for rhythmic movements, the final target is the assistance/rehabilitation of locomotory tasks. As an initial proof of concept, this paper focuses on a simpler movement, i.e. rhythmic oscillations of the forearm about the elbow.

Multidirectional transparent support for overground gait training

Gait and balance training is an essential ingredient for locomotor rehabilitation of patients with neurological impairments. Robotic overhead support systems may help these patients train, for example by relieving them of part of their body weight. However, there are only very few systems that provide support during overground gait, and these suffer from limited degrees of freedom and/or undesired interaction forces due to uncompensated robot dynamics, namely inertia. Here, we suggest a novel mechanical concept that is based on cable robot technology and that allows three-dimensional gait training while reducing apparent robot dynamics to a minimum. The solution does not suffer from the conventional drawback of cable robots, which is a limited workspace. Instead, displaceable deflection units follow the human subject above a large walking area. These deflection units are not actuated, instead they are implicitly displaced by means of the forces in the cables they deflect. This leads to an underactuated design, because the deflection units cannot be moved arbitrarily. However, the design still allows accurate control of a three-dimensional force vector acting on a human subject during gait. We describe the mechanical concept, the control concept, and we show first experimental results obtained with the device, including the force control performance during robot-supported overground gait of five human subjects without motor impairments.

Oscillator-based walking assistance: A model-free approach

In this paper, we further develop our framework to design new assistance and rehabilitation protocols based on motor primitives. In particular, we extend our recent results of oscillator-based assistance to the case of walking. The adaptive oscillator used in this paper is capable of predicting the angular position of the users joints in the future, based on the pattern learned during preceding cycles. Assistance is then provided by attracting the joints to this future position using a force field in a compliant lower-limb exoskeleton. To demonstrate the method efficiency, we computed the rate of metabolic energy expended by the participants during a walking task, with and without assistance. Results show a significant decrease of energy expenditure with the assistance switched on, although not to a point to entirely compensate for the burden due to the exoskeleton lack of transparency. The results further show changes in the kinematics: with assistance, the participants walked with a faster cadence and ampler movements. These results tend to prove the relevance of designing assistance protocols based on adaptive oscillators (or primitives in general) and pave the way to the design of new rehabilitation protocols.

Multisensory Integration in Dynamical Behaviors: Maximum Likelihood Estimation across Bimanual Skill Learning

Optimal integration of different sensory modalities weights each modality as a function of its degree of certainty (maximum likelihood). Humans rely on near-optimal integration in decision-making tasks (involving e.g., auditory, visual, and/or tactile afferents), and some support for these processes has also been provided for discrete sensorimotor tasks. Here, we tested optimal integration during the continuous execution of a motor task, using a cyclical bimanual coordination pattern in which feedback was provided by means of proprioception and augmented visual feedback (AVF, the position of both wrists being displayed as the orthogonal coordinates of a single cursor). Assuming maximum likelihood integration, the following predictions were addressed: (1) the coordination variability with both AVF and proprioception available is smaller than with only one of the two modalities, and should reach an optimal level; (2) if the AVF is artificially corrupted by noise, variability should increase but saturate toward the level without AVF; (3) if the AVF is imperceptibly phase shifted, the stabilized pattern should be partly adapted to compensate for this phase shift, whereby the amount of compensation reflects the weight assigned to AVF in the computation of the integrated signal. Whereas performance variability gradually decreased over 5 d of practice, we showed that these model-based predictions were already observed on the first day. This suggests not only that the performer integrated proprioceptive feedback and AVF online during task execution by tending to optimize the signal statistics, but also that this occurred before reaching an asymptotic performance level.