Narrow Results

We investigated the influence of three different high-pass (HP) and low-pass (LP) filtering conditions and a Gaussian (GNMF) and inverse-Gaussian (IGNMF) non-negative matrix factorization algorithm on the extraction of muscle synergies from myoelectric signals during human walking and running. To evaluate the effects of signal recording and processing on the outcomes, we analyzed the intraday and interday computation reliability. Results show that the IGNMF achieved a significantly higher reconstruction quality and on average needs one less synergy to sufficiently reconstruct the original signals compared to the GNMF. For both factorizations, the HP with a cut-off frequency of 250Hz significantly reduces the number of synergies. We identified the filter configuration of fourth order, HP 50Hz and LP 20Hz as the most suitable to minimize the combination of fundamental synergies, providing a higher reliability across all filtering conditions even if HP 250Hz is excluded. Defining a fundamental synergy as a single-peaked activation pattern, for walking and running we identified five and six fundamental synergies, respectively using both algorithms. The variability in combined synergies produced by different filtering conditions and factorization methods on the same data set suggests caution when attributing a neurophysiological nature to the combined synergies.

The optimal aquatic locomotion has previously been associated with a narrow St(= fA/u) number range of 0.2–0.4. We present how animals tune their Strouhal (St) number to this range to reveal the mechanisms influencing efficiency. The self-propelled swimming of a 2D swimmer is simulated using an immersed boundary method. The locomotion kinematics is controlled by two variables, $A^{\ast }(=A/L)$ and frequency f. We show that only when animals constrain their $A^{\ast }$ = 0.125–0.25, their St number can fall into the optimal St range. When $f>0.4$ Hz, the St number is independent with frequency. Although different combinations of f and $A^{\ast }$ can achieve a same cruising velocity, high-f and low-$A^{\ast }$ motions are more efficient. This can be linked to its larger lateral velocity component in the proto-vortex region and the transition of the tail vortices into small eddies.

Symmetries in the external world constrain the evolution of neuronal circuits that allow organisms to sense the environment and act within it. Many small “modular” circuits can be viewed as approximate discretizations of the relevant symmetries, relating their forms to the functions they perform. The recent development of a formal theory of dynamics and bifurcations of networks of coupled differential equations permits the analysis of some aspects of network behavior without invoking specific model equations or numerical simulations. We review basic features of this theory, compare it to equivariant dynamics, and examine the subtle effects of symmetry when combined with network structure. We illustrate the relation between form and function through examples drawn from neurobiology, including locomotion, peristalsis, visual perception, balance, hearing, location detection, decision-making, and the connectome of the nematode Caenorhabditis elegans.

The behavior of an animated artwork, survivor — a classroom chair which walks, with a dynamics which some viewers find haunting — reflects an attempt to emulate (and suggest to viewers) some feelings and behaviors that are typical of survivors of landmine blasts, learning to use crutches. The artwork itself is intended for sensitizing viewers to the horror experienced by those who survive, and those who do not. The behavior of such a survivor is affected by several factors: some are due to the objective difficulty of using prosthetic legs, and some are due to emotional factors, e.g., fear, "shame" of being in such situation, and pain. The mechanical structure, strongly conditioned by artistic requirements, was combined with a control system that exhibits appropriate behaviors. Behavioral control, a technique developed for the control of mobile robots, was used in survivor, and implemented over a modified version of the traditional Brooks' subsumption architecture. This technique makes it possible to emulate normal locomotion behaviors such as the need of avoiding obstacles and typical animal feelings such as curiosity, hunger, fatigue and fear. We describe the mechanics and viewers' response, and formalize aesthetic response. We briefly survey computer modelling of emotions, robotic art, and biomimetic locomotion in robotics.

We have questioned whether a complex behavior, such as fish swimming, can be better described quantitatively as a sequence of discrete events or states than with classical kinematic measures which can be compromised by inherent variability. Here, the different states, expressed as combinations of symbols, were defined on the basis of the animal's location (A: periphery, and B: inner part of the aquarium) and speed (Fast and Slow). We observed that the distributions of time intervals spent in the successive states were not gaussian. Rather, they were fit by power laws associated with an underlying Lévy-like process which has more long intervals, primarily due to prolonged periods of relative inactivity. Furthermore, our data suggest that the swimming behavior can be attributed to interactions between two intrinsic systems. One is represented by the matrix of transition of probabilities between states and controls their sequential organization while the second, which is defined by interval distributions, determines the time spent in each state. This kinetic model detects subtle effects of low doses of neuroactive compounds, and identifies their specific locus of action. We propose that this paradigm can be applied to characterize normal behavior and its modifications by genetic or pharmacological manipulations.

In this research, we apply complexity-based techniques to study the activations of the brain while the subjects perform different types of locomotion, including walking, jogging, and running. Therefore, we can study the effect of locomotion speed (or toughness level) on brain’s reactions. For this purpose, we analyzed the fractal dimension and approximate entropy of electroencephalogram (EEG) signals recorded from subjects while they walked, jogged, and ran for 20 s in the case of each activity. The analysis of 21 recorded samples showed that the complexity of EEG signals increases by increasing the locomotion speed. This result indicates a higher level of processing in the brain while the subjects perform a harder task. This analysis can be extended to the case of other physiological signals to study the effect of the level of exercise on different organs’ activations.

The purpose of this study was to develop a model to estimate lower extremity joint moments during level gait. A three-layer artificial neural network was developed to map diverse inputs (demographics, anthropometrics, electromyography, kinematics) onto sagittal plane resultant joint moments for a sample of healthy young adults. Overall model performance and prediction accuracy were acceptable for the hip, knee, and ankle, with coefficients of determination (r²) reaching 0.90 for the hip and knee and 0.95 for the ankle. Estimates in the case-specific validation produced r² values of 0.95, 0.94, and 0.99 for the hip, knee, and ankle, respectively. Absolute errors of estimation for peak data were within the ranges published previously for other joint moment models. The results indicated that the model used in this study is accurate in estimating sagittal plane joint moments about the hip, knee, and ankle. Furthermore, the model retained accuracy with a reduced list of inputs (kinematics and demographics). Future development will include clinical samples to determine the adaptability of this model to the diverse conditions of musculoskeletal gait dysfunction common in the clinical setting.

This paper introduces a data-driven approach for human locomotion generation that takes as input a set of example locomotion clips and a motion path specified by an animator. Significantly, the approach only requires a single example of straight-path locomotion for each style expressed and can produce a continuous output sequence on an arbitrary path. Our approach considers quantitative and qualitative aspects of motion and suggests several techniques to synthesize a convincing output animation: motion path generation, interactive editing, and physical enhancement for the output animation. Initiated with an example clip, this process produces motion that differs stylistically from any in the example set, yet preserves the high quality of the example motion. As shown in the experimental results, our approach provides efficient locomotion generation by editing motion capture clips, especially for a novice animator, at interactive speed.

Virtual proprioception represents a novel means of developing cortical reorganization of alternative strategies for hemiparetic gait. Fundamentals of the device are motor control plasticity, aftereffect, and visual-based biofeedback. Two wireless three-dimensional (3D) microelectromechanical systems (MEMS) accelerometers are placed on the femur (upper leg) of both the affected and unaffected limbs above the lateral epicondyle next to the knee joint. The acceleration data from the two wireless 3D MEMS accelerometers are fed back to the user in real time by visual output from a portable laptop PC. Given the virtual proprioception feedback, the user can then adjust the original gait while walking to an improved alternative gait strategy. First, hemiparetic gait is comprehensively discussed. The inherent roles of proprioception with locomotion and issues with traumatic brain injury are considered. Then, the technology advance of accelerometers and gait analysis is detailed. Virtual proprioception is tested and evaluated, while demonstrating the capacity to improve disparities in hemiparetic gait during real time.

Assessment of locomotion quality subsequent to neurological trauma, such as stroke or traumatic brain injury is imperative for the correct allocation of therapy dosage and strategy. In light of the limited amount of medical professionals in contrast with the rising number of people with neurological disorders; a new paradigm for addressing therapeutic strategies for neurological trauma is advocated. An important aspect for therapy of neuro-motor disorders is the characterization of gait. There are devices presently used for evaluating gait, such as EMG, optical sensors, electrogoniometers, metabolic energy expenditure devices, foot stride analyzers, and ground reaction force sensors. These devices have inherent issues, such as spatial constraints, line of sight requirements, and specialization requirements. A solution for improved autonomy of gait assessment is demonstrated by the use of fully wireless 3D MEMS accelerometers, which are light weight and minimally intrusive. To minimize specialization issues, the accelerometers may be positioned at a standard anatomical anchor. The role of traumatic brain injury with respect to gait dysfunction is addressed. Enclosed is the initial test and evaluation of a wireless 3D MEMS accelerometer for gait analysis. The gait analysis is conducted in outdoor conditions, while walking on a sidewalk.

The proper allocation of a therapy strategy and dosage is fundamentally associated with the quantified evaluation of gait quality. Wireless accelerometer systems for the evaluation of quantified hemiplegic gait characteristics has been successfully applied in inherently autonomous environments through the consideration of the temporal domain of the gait acceleration waveform. The frequency domain has notable potential for identifying the quantified disparity of the affected leg and unaffected leg through the application of a tandem-activated wireless accelerometer system mounted to the lateral malleolus of each lower leg through an elastic band. The quantification of disparity for hemiplegic gait via the application of wireless accelerometers was applied in an outdoor environment, while walking on a sidewalk. In addition, the wireless accelerometers were tandem activated while the subject had achieved steady-state gait status, which mitigated the need to subjectively remove starting acceleration and stopping deceleration aspects of the gait cycle. Four predominant frequencies within the 0–5 Hz bandwidth demonstrated a considerable degree of accuracy and reliability. The organization of the four predominant frequencies for both affected leg and unaffected leg were found to be disparate in a statistically significant manner, implicating a disparity of the rhythmicity respective of the affected leg in contrast to the unaffected leg in hemiplegic gait. These preliminary findings may advance gait quantification techniques, which may improve the efficacy of gait rehabilitation therapy. Enclosed are the initial test and evaluation of a tandem-activated wireless accelerometer system using the frequency domain for ascertaining a quantified disparity of hemiplegic gait.

Humanoid robots that have to operate in cluttered and unstructured environments, such as man-made and natural disaster scenarios, require sophisticated sensorimotor capabilities. A crucial prerequisite for the successful execution of whole-body locomotion and manipulation tasks in such environments is the perception of the environment and the extraction of associated environmental affordances, i.e., the action possibilities of the robot in the environment. We believe that such a coupling between perception and action could be a key to substantially increase the flexibility of humanoid robots.

In this paper, we approach the affordance-based generation of whole-body actions for stable locomotion and manipulation. We incorporate a rule-based system to assign affordance hypotheses to visually perceived environmental primitives in the scene. These hypotheses are then filtered using extended reachability maps that carry stability information, for identifying reachable affordance hypotheses. We then formulate the hypotheses in terms of a constrained inverse kinematics problem in order to find whole-body configurations that utilize a chosen set of hypotheses.

The proposed methods are implemented and tested in simulated environments based on RGB-D scans as well as on a real robotic platform.

This paper proposes a novel central pattern generator (CPG) model with proprioceptive mechanism and the dynamic connectivity mechanism. It not only contains the sensory information of the environment but also contains the information of the actuators and automatically tunes the parameters of CPG corresponding to the actuators information and inner sensory information. The position of the joints linked directly with the output of CPG is introduced to the CPG to find its proprioceptive system, spontaneously making the robot realize the actuator working status, further changing the CPG output to fit the change and decrease the influence of the problematic joints or actuators on the robot being controlled. So the damage would be avoided and self-protection is implemented. Its application on the locomotion control of a quadruped robot demonstrates the effectiveness of the proposed approach.

Production of energy is a foundation of life. The metabolic rate of organisms (amount of energy produced per unit time) generally increases slower than organisms’ mass, which has important implications for life organization. This phenomenon, when considered across different taxa, is called interspecific allometric scaling. Its origin has puzzled scientists for many decades, and still is considered unknown. In this paper, we posit that natural selection, as determined by evolutionary pressures, leads to distribution of resources, and accordingly energy, within a food chain, which is optimal from the perspective of stability of the food chain, when each species has sufficient amount of resources for continuous reproduction, but not too much to jeopardize existence of other species. Metabolic allometric scaling (MAS) is then a quantitative representation of this optimal distribution. Taking locomotion and the primary mechanism for distribution of energy, we developed a biomechanical model to find energy expenditures, considering limb length, skeleton mass and speed. Using the interspecific allometric exponents for these three measures and substituting them into the locomotion-derived model for energy expenditure, we calculated allometric exponents for mammals, reptiles, fish, and birds, and compared these values with allometric exponents derived from experimental observations. The calculated allometric exponents were nearly identical to experimentally observed exponents for mammals, and very close for fish, reptiles and the basal metabolic rate (BMR) of birds. The main result of the study is that the MAS is a function of a mechanism of optimal energy distribution between the species of a food chain. This optimized sharing of common resources provides stability of a food chain for a given habitat and is guided by evolutionary pressures and natural selection.

In many applications of human–computer interaction, a prediction of the human’s next intended action is highly valuable. To control direction and orientation of the body when walking towards a goal, a walking person relies on visual input obtained by eye and head movements. The analysis of these parameters might allow us to infer the intended goal of the walker. However, such a prediction of human locomotion intentions is a challenging task, since interactions between these parameters are nonlinear and highly dynamic. We employed machine learning models to investigate if walk and gaze data can be used for locomotor prediction. We collected training data for the models in a virtual reality experiment in which 18 participants walked freely through a virtual environment while performing various tasks (walking in a curve, avoiding obstacles and searching for a target). The recorded position, orientation- and eye-tracking data was used to train an LSTM model to predict the future position of the walker on two different time scales, short-term predictions of 50ms and long-term predictions of 2.5s. The trained LSTM model predicted free walking paths with a mean error of 5.14mm for the short-term prediction and 65.73cm for the long-term prediction. We then investigated how much the different features (direction and orientation of the head and body and direction of gaze) contributed to the prediction quality. For short-term predictions, position was the most important feature while orientation and gaze did not provide a substantial benefit. In long-term predictions, gaze and orientation of the head and body provided significant contributions. Gaze offered the greatest predictive utility in situations in which participants were walking short distances or in which participants changed their walking speed.

Generating good and human-like locomotion or other legged motions for bipedal robots has always been challenging. One of the emerging solutions to this challenge is to use imitation learning. The sources for imitation are mostly state-only demonstrations, so using state-of-the-art Generative Adversarial Imitation Learning (GAIL) with Imitation from Observation (IfO) ability will be an ideal framework to use in solving this problem. However, it is often difficult to allow new or complicated movements as the common sources for these frameworks are either expensive to set up or hard to produce satisfactory results without computationally expensive preprocessing, due to accuracy problems. Inspired by how people learn advanced knowledge after acquiring a basic understanding of specific subjects, this paper proposes a Motion capture-aided Video Imitation (MoVI) learning framework based on Adversarial Motion Priors (AMP) by combining motion capture data of primary actions like walking with video clips of target motion like running, aiming to create smooth and natural imitation results of the target motion. This framework is able to produce various human-like locomotion by taking the most common and abundant motion capture data with any video clips of motion without the need for expensive datasets or sophisticated preprocessing.

Locomotor state transitions are challenging for transfemoral (TF) amputees due to the lack of active knee control even in the current powered prosthetic devices. Myoelectric activation has been used successfully to classify steady-state locomotion states, but classification of transitions between locomotion states remains a challenge, especially for TF amputees. The purpose of this study was to determine if lower-extremity muscle activation differences between pre-transition and transition gait cycles occur in the involved or uninvolved limb of TF amputees during locomotion state transitions. Surface electromyography (EMG) was collected from residual muscles on the involved limb and from the uninvolved limb from five TF amputees as they transitioned between different locomotion states (level ground, ramp ascent/descent, stair ascent/descent). Statistical parametric mapping (SPM) was used to assess differences in activation. When analyzed as a group, the only significant differences were observed in the vastus lateralis of the uninvolved limb. High inter-subject variation reduced the significance of other pattern differences. Further inspection revealed that the individual subjects expressed three different recruitment patterns. These recruitment patterns may indicate compensatory strategies adopted by the subjects over the years since amputation. Furthermore, the separate recruitment patterns suggest the need for individualized locomotion transition classification algorithms rather than a general classification scheme.

In this paper, we analyze the locomotion of a double screw-like robot over compliant surfaces. We first develop an analytical model of locomotion over rigid surface. Then we use this model to determine the forces acting on the robot and environment to define the surface properties that allow the robot locomotion. Finally, we performed experiments using the robot to determine its thrust force and its speed over hollow silicone rubber.

The gallop is the preferred gait by mammals for agile traversal through terrain. This motion is intrinsically complex as the feet are used individually and asymmetrically. Experimental data for the gallop are limited due the large workspace needed because of the gait's speed and long traversal. A generalized motion measurement strategy is adopted based on high-speed, motion capture with a reduced marker set and an emphasis on body and leg kinematics and with limited ground reaction force measurement. This allows for an extension of the workspace and allows for markers to be placed in locations with reduced tissue compliance. This is sufficient for capturing the principal motion and for making kinematic comparisons to a previously developed approximating impulse model framework. A series of gallops were measured in a large gait laboratory (18 m² principal working area) from three canine subjects (ranging from 8 to 24 kg) galloping down a 15 m runway. Normalized results show a correlation with motions suggested by the impulse model and are in keeping with insights from previous animal and legged robot studies.

This article refers to the field of apedal ferrofluid based locomotion systems. An analytical solution shall be given on the matter of flow manipulation by an alternating magnetic field. The problem involves the causative magnetic field leads to the correspondingly deformed free ferrofluid surface contour and presents the evidence of an effective flow, meaning a non-zero average flow rate.