A walking robot called human: lessons to be learned from neural control of locomotion

Augmented Hip Proprioception Influences Mediolateral Foot Placement During Walking

Hip abductor proprioception contributes to the control of mediolateral foot placement, which varies with step-by-step fluctuations in pelvis dynamics. Prior work has used hip abductor vibration as a sensory probe to investigate the link between vibration within a single step and subsequent foot placement. Here, we extended prior findings by applying time and location varying vibration in every step, seeking to predictably manipulate the continuous, step-by-step relationship between pelvis dynamics and foot placement. We compared participants' (n = 32; divided into two groups of 16 with slightly different vibration control) gait behavior across four treadmill walking conditions: 1) No feedback; 2) Random feedback, with vibration unrelated to pelvis motion; 3) Augmented feedback, with vibration designed to evoke proprioceptive feedback paralleling the actual pelvis motion; 4) Disrupted feedback, with vibration designed to evoke proprioceptive feedback inversely related to pelvis motion. We hypothesized that the relationship between pelvis dynamics and foot placement would be strengthened by Augmented feedback but weakened by Disrupted feedback. For both participant groups, the strength of the relationship between pelvis dynamics at the start of a step and foot placement at the end of a step was significantly (p ≤ 0.0002) influenced by the feedback condition. The link between pelvis dynamics and foot placement was strongest with Augmented feedback, but not significantly weakened with Disrupted feedback, partially supporting our hypotheses. Our approach to augmenting proprioceptive feedback during gait may have implications for clinical populations with a weakened relationship between pelvis motion and foot placement.

Human muscle activity and lower limb biomechanics of overground walking at varying levels of simulated reduced gravity and gait speeds

Reducing the mechanical load on the human body through simulated reduced gravity can reveal important insight into locomotion biomechanics. The purpose of this study was to quantify the effects of simulated reduced gravity on muscle activation levels and lower limb biomechanics across a range of overground walking speeds. Our overall hypothesis was that muscle activation amplitudes would not decrease proportionally to gravity level. We recruited 12 participants (6 female, 6 male) to walk overground at 1.0, 0.76, 0.55, and 0.31 G for four speeds: 0.4, 0.8, 1.2, and 1.6 ms^-1. We found that peak ground reaction forces, peak knee extension moment in early stance, peak hip flexion moment, and peak ankle extension moment all decreased substantially with reduced gravity. The peak knee extension moment at late stance/early swing did not change with gravity. The effect of gravity on muscle activity amplitude varied considerably with muscle and speed, often varying nonlinearly with gravity level. Quadriceps (rectus femoris, vastus lateralis, & vastus medialis) and medial gastrocnemius activity decreased in stance phase with reduced gravity. Soleus and lateral gastrocnemius activity had no statistical differences with gravity level. Tibialis anterior and biceps femoris increased with simulated reduced gravity in swing and stance phase, respectively. The uncoupled relationship between simulated gravity level and muscle activity have important implications for understanding biomechanical muscle functions during human walking and for the use of bodyweight support for gait rehabilitation after injury.

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.

Walking with perturbations: a guide for biped humans and robots

This paper provides an update on the neural control of bipedal walking in relation to bioinspired models and robots. It is argued that most current models or robots are based on the construct of a symmetrical central pattern generator (CPG). However, new evidence suggests that CPG functioning is basically asymmetrical with its flexor half linked more tightly to the rhythm generator. The stability of bipedal gait, which is an important problem for robots and biological systems, is also addressed. While it is not possible to determine how biological biped systems guarantee stability, robot solutions can be useful to propose new hypotheses for biology. In the second part of this review, the focus is on gait perturbations, which is an important topic in robotics in view of the frequent falls of robots when faced with perturbations. From the human physiology it is known that the initial reaction often consists of a brief interruption followed by an adequate response. For instance, the successful recovery from a trip is achieved using some basic reactions (termed elevating and lowering strategies), that depend on the phase of the step cycle of the trip occurrence. Reactions to stepping unexpectedly in a hole depend on comparing expected and real feedback. Implementation of these ideas in models and robotics starts to emerge, with the most advanced robots being able to learn how to fall safely and how to deal with complicated disturbances such as provided by walking on a split-belt.

Bipedal robotic walking control derived from analysis of human locomotion

This paper proposes the design of a bipedal robotic controller where the function between the sensory input and motor output is treated as a black box derived from human data. In order to achieve this, we investigated the causal relationship between ground contact information from the feet and leg muscle activity n human walking and calculated filter functions which transform sensory signals to motor actions. A minimal, nonlinear, and robust control system was created and subsequently analysed by applying it to our bipedal robot RunBot III without any central pattern generators or precise trajectory control. The results demonstrate that our controller can generate stable robotic walking. This indicates that complex locomotion patterns can result from a simple model based on reflexes and supports the premise that human-derived control strategies have potential applications in robotics or assistive devices.

Locomotor Sub-functions for Control of Assistive Wearable Robots

A primary goal of comparative biomechanics is to understand the fundamental physics of locomotion within an evolutionary context. Such an understanding of legged locomotion results in a transition from copying nature to borrowing strategies for interacting with the physical world regarding design and control of bio-inspired legged robots or robotic assistive devices. Inspired from nature, legged locomotion can be composed of three locomotor sub-functions, which are intrinsically interrelated: Stance: redirecting the center of mass by exerting forces on the ground. Swing: cycling the legs between ground contacts. Balance: maintaining body posture. With these three sub-functions, one can understand, design and control legged locomotory systems with formulating them in simpler separated tasks. Coordination between locomotor sub-functions in a harmonized manner appears then as an additional problem when considering legged locomotion. However, biological locomotion shows that appropriate design and control of each sub-function simplifies coordination. It means that only limited exchange of sensory information between the different locomotor sub-function controllers is required enabling the envisioned modular architecture of the locomotion control system. In this paper, we present different studies on implementing different locomotor sub-function controllers on models, robots, and an exoskeleton in addition to demonstrating their abilities in explaining humans' control strategies.

Reflex control of robotic gait using human walking data

Control of human walking is not thoroughly understood, which has implications in developing suitable strategies for the retraining of a functional gait following neurological injuries such as spinal cord injury (SCI). Bipedal robots allow us to investigate simple elements of the complex nervous system to quantify their contribution to motor control. RunBot is a bipedal robot which operates through reflexes without using central pattern generators or trajectory planning algorithms. Ground contact information from the feet is used to activate motors in the legs, generating a gait cycle visually similar to that of humans. Rather than developing a more complicated biologically realistic neural system to control the robot's stepping, we have instead further simplified our model by measuring the correlation between heel contact and leg muscle activity (EMG) in human subjects during walking and from this data created filter functions transferring the sensory data into motor actions. Adaptive filtering was used to identify the unknown transfer functions which translate the contact information into muscle activation signals. Our results show a causal relationship between ground contact information from the heel and EMG, which allows us to create a minimal, linear, analogue control system for controlling walking. The derived transfer functions were applied to RunBot II as a proof of concept. The gait cycle produced was stable and controlled, which is a positive indication that the transfer functions have potential for use in the control of assistive devices for the retraining of an efficient and effective gait with potential applications in SCI rehabilitation.

A physical model of sensorimotor interactions during locomotion

In this paper, we describe the development of a bipedal robot that models the neuromuscular architecture of human walking. The body is based on principles derived from human muscular architecture, using muscles on straps to mimic agonist/antagonist muscle action as well as bifunctional muscles. Load sensors in the straps model Golgi tendon organs. The neural architecture is a central pattern generator (CPG) composed of a half-center oscillator combined with phase-modulated reflexes that is simulated using a spiking neural network. We show that the interaction between the reflex system, body dynamics and CPG results in a walking cycle that is entrained to the dynamics of the system. We also show that the CPG helped stabilize the gait against perturbations relative to a purely reflexive system, and compared the joint trajectories to human walking data. This robot represents a complete physical, or 'neurorobotic', model of the system, demonstrating the usefulness of this type of robotics research for investigating the neurophysiological processes underlying walking in humans and animals.

Stepping with an ankle foot orthosis re-examined: a mechanical perspective for clinical decision making

BACKGROUND

Ankle foot orthoses are used to stabilize the ankle joint and aid toe clearance during stepping in persons after incomplete spinal cord injury. However, little is known about kinematics and kinetics of stepping with an orthosis during the transition from stance-to-swing and swing-to-stance. We intended to examine if an ankle foot orthosis impeded or facilitated optimal ankle, knee and hip joint kinematics, kinetics and spatiotemporal parameters during the transition phases of normal walking.

METHODS

Fourteen healthy participants walked on a split-belt instrumented treadmill with and without a posterior leaf spring ankle foot orthosis at 1.2m/s. Three dimensional motion data and ground reaction forces were captured during 30-second trials of steady state walking.

FINDINGS

During stance-to-swing, the orthosis significantly decreased hip extension [8.6 (5.5) to 6.7 (5.5) degrees, P=0.001], ankle plantarflexion [19.4 (5.7) to 12.0 (5.2) degrees, P<0.001] and plantarflexor power [0.18 (0.03) to 0.15 (0.03) watts/body weight, P<0.001]. During swing-to-stance, the orthosis significantly increased hip flexion [32.7 (4.7) to 35.6 (5.1) degrees, P=0.028] and ankle plantarflexion [8.4 (3.5) to 10.9 (4.7) degrees, P<0.001] and decreased loading rate [0.06 (0.01) to 0.05 (0.01) N/kg, P=0.018] and braking force [0.16 (0.02) to 0.15 (0.02) N/kg, P=0.013]. Double limb support time increased significantly with the orthosis [0.19 (0.02) to 0.22 (0.03) seconds, P<0.000].

INTERPRETATION

An ankle foot orthosis affected joint kinematics and kinetics during the transition from stance-to-swing and vice-versa. The use of orthosis to improve transition phase kinematics and kinetics in individuals with incomplete spinal cord injury warrants assessment.

Energy expenditure during human gait. II - Role of muscle groups

A phenomenological model of muscle energy expenditure developed in part I of the paper, is utilized as a physiological cost function to estimate the muscle forces during normal locomotion. The model takes into account muscular behaviors typically observed during human gait, such as submaximal activation, variable muscular contraction conditions and muscular fiber type. The solution of the indeterminate biomechanical problem is obtained by integrating multibody dynamics and the global static optimization technique that considers the whole motion. The results for an application case indicate the important role of muscle groups in coordinating multijoint motion with the objective of minimizing metabolic costs of transport during locomotion.

Growth of segment parameters and a morphological classification for children between 15 and 36 months

This study is part of a research program that aims at a better understanding of the influence of individual morphological differences and physical growth on development of independent walking in toddlers. As morphometric and segment inertial parameters for toddlers aged between 15 and 36 months are indispensable for the mechanical analyses inherent to this purpose, parameter data were collected. The provided dataset of morphological and segment inertial parameters is a valuable tool for locomotor biomechanical modelling. Analysis of the parameter data showed that there are substantial changes of most segment inertial parameters across body length and body mass. In addition, a classification system was developed to categorize toddlers on the basis of morphometry, reflecting the segment inertial constitution of the child. A principal components analysis (PCA) was applied to define the variance in physique between the children. PCA resulted in three newly composed variables: the 'Axis of chubbiness', the 'Axis of allometric growth' and the 'Axis of relative limb length'. The three axes are plotted against each other, resulting in eight morphological classes. With this classification the morphotype of toddlers between 15 and 36 months can be specified and used for further research on their walking patterns.

Influence of falling height on the excitability of the soleus H-reflex during drop-jumps

AIM

The stretch-shortening cycle (SSC) is characterized by stretching of the target muscle (eccentric phase) prior to a subsequent shortening in the concentric phase. Stretch reflexes in the eccentric phase were argued to influence the performance of short lasting SSCs. In drop-jumps, the short latency component of the stretch reflex (SLR) was shown to increase with falling height. However, in jumps from excessive heights, the SLR was diminished. So far, it is unclear whether the modulation of the SLR relies on spinal mechanisms or on an altered fusimotor drive. The present study aimed to assess the spinal excitability of the soleus Ia afferent pathway at SLR during jumps from low height (LH - 31 cm) and excessive height (EH - 76 cm).

METHODS

In 20 healthy subjects (age 25 +/- 3 years), H-reflexes were timed to occur at the peak of the SLR during drop-jumps from LH and EH.

RESULTS

H-reflexes were significantly smaller at EH than at LH (P < 0.05). Neither soleus and tibialis anterior background EMG nor the size of the maximum M-wave changed with falling height.

CONCLUSION

Differences in the H-reflex between EH and LH indicate that spinal mechanisms are involved in the modulation of the SLR. A decreased excitability of the H-reflex pathway at EH compared with LH is argued to serve as a 'prevention strategy' to protect the tendomuscular system from potential injuries caused by the high load. It is argued that pre-synaptic inhibition of Ia afferents is most likely responsible for the change in H-reflex excitability between the two jump conditions.

Changes in 3D joint dynamics during the first 5 months after the onset of independent walking: a longitudinal follow-up study

A longitudinal follow-up study of 10 normally developing children was performed in order to identify changes in mechanical control of gait during the first months after initiation of independent walking. Changes in spatio-temporal parameters and kinematics were recorded (336 trials spread over 83 recording sessions) and linked to kinetic features of gait. At the onset of independent walking, all children in our study group showed the same walking strategy: a dominance of the extending moments around the lower extremity joints was observed and could be linked to the flexed position of the hip and knee during stance. In a subset of our study population, the dominance of the extending moments disappeared with increasing walking experience, though reversal to immature patterns was frequently observed. A linear mixed model showed that with increasing walking experience, there was an increase in dimensionless walking speed, dimensionless cadence and dimensionless stride length (without correction for the increase in speed). Maximal hip extension in stance, knee flexion and ankle plantar flexion at foot contact also increased (even when the increase in speed is taken into account). Dimensionless step width, duty factor, double support time, maximal hip flexion in swing and hip abduction significantly decreased (with correction for speed). Important changes were also observed in ground reaction force patterns, evolving towards a double "hump". No significant changes could be observed in other kinetic parameters, probably due to the high degree of variability.

Neuroplasticity after spinal cord injury and training: an emerging paradigm shift in rehabilitation and walking recovery

Physical rehabilitation after spinal cord injury has been based on the premise that the nervous system is hard-wired and irreparable. Upon this assumption, clinicians have compensated for irremediable sensorimotor deficits using braces, assistive devices, and wheelchairs to achieve upright and seated mobility. Evidence from basic science, however, demonstrates that the central nervous system after injury is malleable and can learn, and this evidence has challenged our current assumptions. The evidence is especially compelling concerning locomotion. The purpose of this perspective article is to summarize the evidence supporting an impending paradigm shift from compensation for deficits to rehabilitation as an agent for walking recovery. A physiologically based approach for the rehabilitation of walking has developed, translating evidence for activity-dependent neuroplasticity after spinal cord injury and the neurobiological control of walking. Advanced by partnerships among neuroscientists, clinicians, and researchers, critical rehabilitation concepts are emerging for activity-based therapy to improve walking recovery, with promising clinical findings.

Effect of skin movement on the analysis of hindlimb kinematics during treadmill locomotion in rats

In rat gait kinematics, the method most frequently used for measuring hindlimb movement involves placing markers on the skin surface overlying the joints being analyzed. Soft tissue movement around the knee joint has been considered the principle source of error when estimating hindlimb joint kinematics in rodents. However, the motion of knee marker was never quantified, nor the different variations in joint angle associated with this gait analysis system. The purpose of this study was two-fold. The first purpose was to expand upon the limited pool of information describing the effect of soft tissue movement over the knee upon the angular positions of the hip, knee and ankle of rats during treadmill locomotion. Secondly, it was a goal of this study to document the magnitude of the skin displacement when using markers that were attached superficially to the knee joint. This was examined by comparing the hindlimb kinematics in sagittal plane during treadmill locomotion determined from the marker attached to the knee and when the knee position was determined indirectly by computer analysis. Results showed that there is a considerable skin movement artefact which propagates to knee joint position and hindlimb kinematics estimates. It was concluded that these large errors can decrease data reliability in the research of rat gait analysis.