Anthropomorphic robotics

An important factor in trying to capture the complexity of many manipulation problems is the notion of Output Motor Impedance, i.e., the relationship between a set of disturbing forces and the resulting variation in arm configuration. The functional significance of such force/displacement characteristics is investigated, showing how several aspects of different manipulation tasks (holding against gravity, inserting, fast moving, and throwing) can be naturally described in terms of appropriate modulation of the impedance characteristics of the manipulator. For this reason, impedance modulation can be considered an integral part of motor control.     

Stabilization strategies for unstable dynamics

The stabilization of the human standing posture was originally attributed to the stiffness of the ankle muscles but direct measurements of the ankle stiffness ruled out this hypothesis, leaving open the possibility for a feedback stabilization strategy driven by proprioceptive signals. This solution, however, could be implemented with two different kinds of control mechanisms, namely continuous or intermittent feedback. The debate is now settled and the latter solution seems to be the most plausible one. Moreover, stabilization of unstable dynamics is not limited to bipedal standing. Indeed many manipulation tasks can be described in the same framework and thus a very general protocol for addressing this kind of problems is the use of haptic virtual reality where instability is generated by some kind of divergent or saddle-like force field. Several studies demonstrated that human subjects can choose to adopt a stiffness or feedback strategy as a combination of biomechanical and task constraints and can learn to switch from one strategy to the other if it is feasible or to use one or the other is infeasible. Understanding such mechanisms is relevant, for example, for the design of novel ergonomic man-machine interfaces in difficult, unstable tasks.     

Robot-based methodology for a kinematic and kinetic analysis of unconstrained,but reproducible upper extremity movement

Although arm movements play an important role in everyday life, there is still a lack of procedures for the analysis of upper extremity movement. The main problems for standardizing the procedure are the variety of arm movements and the difficult assessment of external hand forces. The first problem requires the predefinition of motions, and the second one is the prerequisite for calculation of net joint forces and torques arising during motion. A new methodology for measuring external forces during prespecified, reproducible upper extremity movement has been introduced and validated. A robot-arm has been used to define the motion and 6 degrees of freedom (DoF) force sensor has been attached to it for acquiring the external loads acting on the arm. Additionally, force feedback has been used to help keeping external loads constant. Intra-individual reproducibility of joint angles was estimated by using correlation coefficients to compare a goal-directed movement with robot-guided task. Inter-individual reproducibility has been evaluated by using the mean standard deviation of joint angles for both types of movement. The results showed that both inter- and intra-individual reproducibility have significantly improved by using the robot. Also, the effectiveness of using force feedback for keeping a constant external load has been shown. This makes it possible to estimate net joint forces and torques which are important biomechanical information in motion analysis.     

Single beam optical trapping integrated in a confocal microscope for biological applications.

Confocal microscopy is very useful in biology because of its three dimensional imaging capacities and has proven to be an excellent tool to study the 3D organization of, for instance, cell structures. This property of confocal microscopy makes it also very suitable for observation during guidance of the three dimensional manipulation of single cells or cell elements. Therefore we decided to integrate a confocal microscope and a single beam optical manipulator into a single instrument. The advantage of optical manipulation over mechanical techniques is that it is non-invasive and therefore may be applied on living (micro-) organisms and cells. The creation of an effective single beam optical trap requires the use of a high numerical aperture (N.A.) objective to focus the laser beam. In this paper we briefly discuss the vertical or axial force exerted on a sphere in a single beam trap. The axial force on a sphere placed on the optical axis, caused by reflection and refraction, is calculated applying a electromagnetic vector diffraction theory to determine the field distribution in the focal region. One of the results is that the particle also experiences a vertical trapping force towards the focusing lens when it is in the strongly convergent part of the field in addition to the known negative signed trapping force in the divergent part of the field. Further we describe an instrumental approach to realize optical trapping in which the optical trap position is controlled by moving the focusing objective only.(ABSTRACT TRUNCATED AT 250 WORDS)     

Winching up heavy loads with a compliant arm: a new local joint controller

A closed kinematic chain, like an arm that operates a crank, has a constrained movement space. A meaningful movement of the chain’s endpoint is only possible along the free movement directions which are given implicitly by the contour of the object that confines the movement of the chain. Many technical solutions for such a movement task, in particular those used in robotics, use central controllers and force–torque sensors in the arm’s wrist or a leg’s ankle to construct a coordinate system (task frame formalism) at the local point of contact the axes of which coincide with the free and constrained movement directions. Motivated by examples from biology, we introduce a new control system that solves a constrained movement task. The control system is inspired by the control architecture that is found in stick insects like Carausius morosus. It consists of decentral joint controllers that work on elastic joints (compliant manipulator). The decentral controllers are based on local positive velocity feedback (LPVF). It has been shown earlier that LPVF enables contour following of a limb in a compliant motion task without a central controller. In this paper we extend LPVF in such a way that it is even able to follow a contour if a considerable counter force drags the limb away along the contour in a direction opposite to the desired. The controller extension is based on the measurement of the local mechanical power generated in the elastic joint and is called power-controlled relaxation LPVF. The new control approach has the following advantages. First, it still uses local joint controllers without knowledge of the kinematics. Second, it does not need a force or torque measurement at the end of the limb. In this paper we test power-controlled relaxation LPVF on a crank turning task in which a weight has to be winched up by a two-joint compliant manipulator.     

Effects of coupled upper limbs movements on postural stabilisation

The preference for in-phase association of coupled cyclic limbs movements is well described (mirror-symmetrical patterns) and this is demonstrated by the ease of performing in-phase movements compared to anti-phase ones. The hypothesis of this study is that the easiest movement patterns are those with minor postural activity. The aim of this study was to describe postural activity in standing subjects in the sagittal and frontal planes during the execution of three upper limbs tasks (single arm, in-phase, anti-phase) at four different frequencies (from 0.6 to 1.2 Hz).We employed six infrared cameras for recording kinematics information, a force platform for measuring forces exerted on the ground, and a system for surface electromyography (SEMG). Outcome measures were: upper limb range of movement and relative-phase, centre of pressure displacement (COP), screw torque (T_z) exerted on the ground, and SEMG recordings of postural muscles (adductor longus, gluteus medius, rectus femoris, and biceps femoris).Our results show that in both the planes the in-phase task resulted in less COP displacement, torque production, and postural muscles involvement than the anti-phase and single arm tasks. This reduced need of postural control could explain the ease of performing in-phase coupled limb movements compared with anti-phase movements.     

Use of Self-Selected Postures to Regulate Multi-Joint Stiffness During Unconstrained Tasks

Background

The human motor system is highly redundant, having more kinematic degrees of freedom than necessary to complete a given task. Understanding how kinematic redundancies are utilized in different tasks remains a fundamental question in motor control. One possibility is that they can be used to tune the mechanical properties of a limb to the specific requirements of a task. For example, many tasks such as tool usage compromise arm stability along specific directions. These tasks only can be completed if the nervous system adapts the mechanical properties of the arm such that the arm, coupled to the tool, remains stable. The purpose of this study was to determine if posture selection is a critical component of endpoint stiffness regulation during unconstrained tasks.

Methodology/Principal Findings

Three-dimensional (3D) estimates of endpoint stiffness were used to quantify limb mechanics. Most previous studies examining endpoint stiffness adaptation were completed in 2D using constrained postures to maintain a non-redundant mapping between joint angles and hand location. Our hypothesis was that during unconstrained conditions, subjects would select arm postures that matched endpoint stiffness to the functional requirements of the task. The hypothesis was tested during endpoint tracking tasks in which subjects interacted with unstable haptic environments, simulated using a 3D robotic manipulator. We found that arm posture had a significant effect on endpoint tracking accuracy and that subjects selected postures that improved tracking performance. For environments in which arm posture had a large effect on tracking accuracy, the self-selected postures oriented the direction of maximal endpoint stiffness towards the direction of the unstable haptic environment.

Conclusions/Significance

These results demonstrate how changes in arm posture can have a dramatic effect on task performance and suggest that postural selection is a fundamental mechanism by which kinematic redundancies can be exploited to regulate arm stiffness in unconstrained tasks.     

Prediction of handgrip forces using surface EMG of forearm muscles.

Evaluation of handgrip forces constitutes an essential component of ergonomic evaluation (e.g. of hand tools), but is difficult to perform at the workplace. The present study describes a series of experiments on 8 healthy male subjects to determine the validity of linear regression models using the surface electromyography (EMG) of up to 6 forearm muscles to predict handgrip forces. For isometric gripping tasks, normalized EMG to grip force calibrations using a series of dynamic force bursts up to 300 N resulted in a valid prediction of grip forces based on the EMG of 6 forearm muscles. Absolute differences between observed and predicted grip force were small (between 27 and 41 N) which shows that the proposed method might be used for the ergonomic evaluation of the use of hand tools. The EMG - handgrip force model appeared to be minimally affected by grip width, i.e. a model for 67 mm grip width was able to validly predict grip forces for 59 and 75 mm grip widths. Furthermore, it was shown that of the 6 forearm muscles studied at least 3 have to be assessed to arrive at a sufficient level of validity, while it seems to be irrelevant which 3 of those 6 forearm muscles are assessed.     

A model for learning human reaching movements

 Reaching movement is a fast movement towards a given target. The main characteristics of such a movement are straight path and a bell-shaped speed profile. In this work a mathematical model for the control of the human arm during ballistic reaching movements is presented. The model of the arm contains a 2 degrees of freedom planar manipulator, and a Hill-type, non-linear mechanical model of six muscles. The arm model is taken from the literature with minor changes. The nervous system is modeled as an adjustable pattern generator that creates the control signals to the muscles. The control signals in this model are rectangular pulses activated at various amplitudes and timings, that are determined according to the given target. These amplitudes and timings are the parameters that should be related to each target and initial conditions in the workspace. The model of the nervous system consists of an artificial neural net that maps any given target to the parameter space of the pattern generator. In order to train this net, the nervous system model includes a sensitivity model that transforms the error from the arm end-point coordinates to the parameter coordinates. The error is assessed only at the termination of the movement from knowledge of the results. The role of the non-linearity in the muscle model and the performance of the learning scheme are analysed, illustrated in simulations and discussed. The results of the present study demonstrate the central nervous system’s (CNS) ability to generate typical reaching movements with a simple feedforward controller that controls only the timing and amplitude of rectangular excitation pulses to the muscles and adjusts these parameters based on knowledge of the results. In this scheme, which is based on the adjustment of only a few parameters instead of the whole trajectory, the dimension of the control problem is reduced significantly. It is shown that the non-linear properties of the muscles are essential to achieve this simple control. This conclusion agrees with the general concept that motor control is the result of an interaction between the nervous system and the musculoskeletal dynamics. Received : 21 May 1996 / Accepted in revised form : 10 June 1997     

An octopus-bioinspired solution to movement and manipulation for soft robots

Soft robotics is a challenging and promising branch of robotics. It can drive significant improvements across various fields of traditional robotics, and contribute solutions to basic problems such as locomotion and manipulation in unstructured environments. A challenging task for soft robotics is to build and control soft robots able to exert effective forces. In recent years, biology has inspired several solutions to such complex problems. This study aims at investigating the smart solution that the Octopus vulgaris adopts to perform a crawling movement, with the same limbs used for grasping and manipulation. An ad hoc robot was designed and built taking as a reference a biological hypothesis on crawling. A silicone arm with cables embedded to replicate the functionality of the arm muscles of the octopus was built. This novel arm is capable of pushing-based locomotion and object grasping, mimicking the movements that octopuses adopt when crawling. The results support the biological observations and clearly show a suitable way to build a more complex soft robot that, with minimum control, can perform diverse tasks.     

A Neuro-Muscular Elasto-Dynamic Model of the Human Arm Part 2: Musculotendon Dynamics and Related Stress Effects

In this paper we develop an elasto-dynamic model of the human arm that includes effects of neuro-muscular control uponelastic deformation in the limb.The elasto-dynamic model of the arm is based on hybrid parameter multiple body systemvariational projection principles presented in the companion paper.Though the technique is suitable for detailed bone and jointmodeling,we present simulations for simplified geometry of the bones,discretized as Rayleigh beams with elongation,whileallowing for large deflections.Motion of the upper extremity is simulated by incorporating muscle forces derived from aHill-type model of musculotendon dynamics.The effects of muscle force are modeled in two ways.In one approach,aneffective joint torque is calculated by multiplying the muscle force by a joint moment ann.A second approach models themuscle as acting along a straight line between the origin and insertion sites of the tendon.Simple arm motion is simulated byutilizing neural feedback and feedforward control.Simulations illustrate the combined effects of neural control strategies,models of muscle force inclusion,and elastic assumptions on joint trajectories and stress and strain development in the bone andtendon.     

Validation of the Delft Shoulder and Elbow Model using in-vivo glenohumeral joint contact forces

The Delft Shoulder and Elbow Model (DSEM), a large-scale musculoskeletal model, is used for the estimation of muscle and joint reaction forces in the shoulder and elbow complex. Although the model has been qualitatively verified using EMG-signals, quantitative validation has until recently not been feasible. The development of an instrumented shoulder endoprosthesis has now made this possible. To this end, motion data, EMG-signals, external forces, and in-vivo glenohumeral joint reaction forces (GH-JRF) were recorded for two patients with an instrumented shoulder hemi-arthroplasty, during dynamic tasks (including abduction and anteflexion) and force tasks with the arm held in a static position. Motions and external forces served as the model inputs to estimate the GH-JRF. In the modeling process, the effect of two different (stress and energy) optimization cost functions and uniform size and mass scaling were evaluated. The model-estimated GH-JRF followed the in-vivo measured force for dynamic tasks up to about 90° arm elevations, but generally underestimates the peak forces up to 31%; whereas a different behavior (ascending measured but descending estimated force) was found for angles above 90°. For the force tasks the model generally overestimated the peak GH-JRF for most directions (on average up to 34%). Applying the energy cost function improved model predictions for the dynamic anteflexion task (up to 9%) and for the force task (on average up to 23%). Scaling also led to improvement of the model predictions during the dynamic tasks (up to 26%), but had a negligible effect (<2%) on the force task results. Although results indicated a reasonable compatibility between model and measured data, adjustments will be necessary to individualize the generic model with the patient-specific characteristics.     

Repeated training with augmentative vibrotactile feedback increases object manipulation performance

Most users of prosthetic hands must rely on visual feedback alone, which requires visual attention and cognitive resources. Providing haptic feedback of variables relevant to manipulation, such as contact force, may thus improve the usability of prosthetic hands for tasks of daily living. Vibrotactile stimulation was explored as a feedback modality in ten unimpaired participants across eight sessions in a two-week period. Participants used their right index finger to perform a virtual object manipulation task with both visual and augmentative vibrotactile feedback related to force. Through repeated training, participants were able to learn to use the vibrotactile feedback to significantly improve object manipulation. Removal of vibrotactile feedback in session 8 significantly reduced task performance. These results suggest that vibrotactile feedback paired with training may enhance the manipulation ability of prosthetic hand users without the need for more invasive strategies.     

Effect of eccentric exercise on position sense at the human forearm in different postures.

This is a study of the ability of blindfolded human subjects to match the position of their forearms before and after eccentric exercise. The hypothesis tested was that the sense of effort contributed to forearm position sense. The fall in force after the exercise was predicted to alter the relationship between effort and force and thereby induce position errors. In the arms-in-front posture, subjects had their unsupported reference arm set to one of two angles from the horizontal, 30 or 60 degrees , and they matched its position by voluntary placement of their other arm. Matching errors were compared with a task where the arms were counterweighted, so could be moved in the vertical plane with minimal effort, and where the arms were moved in the horizontal plane. In these latter two tasks, the intention was to test whether removal of an effort sensation from holding the arm against gravity influenced matching performance. It was found that, although absolute errors for counterweighted and horizontal matching were no larger than for unsupported matching, their standard deviations, 6.1 and 6.8 degrees , respectively, were significantly greater than for unsupported matching (4.6 degrees ), indicating more erratic matching. The eccentric exercise led, the next day, to a fall in maximum voluntary muscle torque of >or=15%. This was accompanied by a significant increase in matching errors for the unsupported matching task from 2.7 +/- 0.5 to 0.8 +/- 0.7 degrees but not for counterweighted (1.4 +/- 0.2 to -0.2 degrees +/- 1.1 degrees ) or horizontal matching (-1.3 +/- 0.7 degrees to -1.8 +/- 0.7 degrees ). This, it is postulated, is because the reduced voluntary torque after exercise was accompanied by a greater effort required to support the arms, leading to larger matching errors. However, effort is only able to provide positional information for unsupported matching where gravity plays a role. In gravity-neutral tasks like counterweighted or horizontal matching, a change in the effort-force relationship after exercise leaves matching accuracy unaffected.     

Activation of shoulder muscles in healthy men and women under isometric conditions

Purpose: Due to the low osseous lead of the shoulder joint a large portion of the shoulder muscles, in addition to executing movements, deals with stabilising tasks. This requires a permanent readjustment of the intermuscular co-ordination of all involved muscles. The aim of the study was to verify the existence of gender dependent differences in intramuscular co-ordination patterns of shoulder muscles.

Method: Fifteen healthy men and nine healthy women, who executed 24 isometric exercises in sagittal, frontal and horizontal planes with a loading of 50% of their individual isometric maximum force, were investigated. In every plane, four angular positions were chosen and both opposite force directions were measured, respectively. SEMG was taken from 13 muscles of the shoulder and the upper arm. Due to inter-individual differences SEMG amplitudes were normalised. Results: Gender specific differences of functional intermuscular co-ordination patterns could be proven systematically. Women showed less activation of muscles acting in the main force direction. In addition, those muscles less necessary for the actual force production were more activated in women than in men.

Conclusions: Functionally comparable shoulder function showed a gender dependency in terms of functional intermuscular co-ordination.     

Effects of Transcranial Direct Current Stimulation on the Control of Finger Force during Dexterous Manipulation in Healthy Older Adults

The contribution of poor finger force control to age-related decline in manual dexterity is above and beyond ubiquitous behavioral slowing. Altered control of the finger forces can impart unwanted torque on the object affecting its orientation, thus impairing manual performance. Anodal transcranial direct current stimulation (tDCS) over primary motor cortex (M1) has been shown to improve the performance speed on manual tasks in older adults. However, the effects of anodal tDCS over M1 on the finger force control during object manipulation in older adults remain to be fully explored. Here we determined the effects of anodal tDCS over M1 on the control of grip force in older adults while they manipulated an object with an uncertain mechanical property. Eight healthy older adults were instructed to grip and lift an object whose contact surfaces were unexpectedly made more or less slippery across trials using acetate and sandpaper surfaces, respectively. Subjects performed this task before and after receiving anodal or sham tDCS over M1 on two separate sessions using a cross-over design. We found that older adults used significantly lower grip force following anodal tDCS compared to sham tDCS. Friction measured at the finger-object interface remained invariant after anodal and sham tDCS. These findings suggest that anodal tDCS over M1 improved the control of grip force during object manipulation in healthy older adults. Although the cortical networks for representing objects and manipulative actions are complex, the reduction in grip force following anodal tDCS over M1 might be due to a cortical excitation yielding improved processing of object-specific sensory information and its integration with the motor commands for production of manipulative forces. Our findings indicate that tDCS has a potential to improve the control of finger force during dexterous manipulation in older adults.     

Neuromuscular dysfunction with the experimental arm acting as its own reference following eccentric and isometric exercise

Eccentric exercise has been extensively used as a model to study muscle damage-induced neuromuscular impairment, adopting mainly a bilateral matching task between the reference (unexercised) arm and the indicator (exercised) arm. However, little attention has been given to the muscle proprioceptive function when the exercised arm acts as its own reference. This study investigated muscle proprioception and motor control, with the arm acting both as reference and indicator, following eccentric exercise and compared them with those observed after isometric exercise. Fourteen young male volunteers were equally divided into two groups and performed an eccentric or isometric exercise protocol with the elbow flexors of the non-dominant arm on an isokinetic dynamometer. Both exercise protocols induced significant changes in indicators of muscle damage, that is, muscle soreness, range of motion and maximal isometric force post-exercise (p < 0.05–0.001), and neuromuscular function was similarly affected following both protocols. Perception of force was impaired over the 4-day post-exercise period (p < 0.001), with the applied force being systematically overestimated. Perception of joint position was significantly disturbed (i.e., target angle was underestimated) only at one elbow angle on day 4 post-exercise (p < 0.05). The misjudgements and disturbed motor output observed when the exercised arm acted as its own reference concur with the view that they could be a result of a mismatch between the central motor command and an impaired motor control after muscle damage.     

Rhythm Patterns Interaction - Synchronization Behavior for Human-Robot Joint Action

Interactive behavior among humans is governed by the dynamics of movement synchronization in a variety of repetitive tasks. This requires the interaction partners to perform for example rhythmic limb swinging or even goal-directed arm movements. Inspired by that essential feature of human interaction, we present a novel concept and design methodology to synthesize goal-directed synchronization behavior for robotic agents in repetitive joint action tasks. The agents’ tasks are described by closed movement trajectories and interpreted as limit cycles, for which instantaneous phase variables are derived based on oscillator theory. Events segmenting the trajectories into multiple primitives are introduced as anchoring points for enhanced synchronization modes. Utilizing both continuous phases and discrete events in a unifying view, we design a continuous dynamical process synchronizing the derived modes. Inverse to the derivation of phases, we also address the generation of goal-directed movements from the behavioral dynamics. The developed concept is implemented to an anthropomorphic robot. For evaluation of the concept an experiment is designed and conducted in which the robot performs a prototypical pick-and-place task jointly with human partners. The effectiveness of the designed behavior is successfully evidenced by objective measures of phase and event synchronization. Feedback gathered from the participants of our exploratory study suggests a subjectively pleasant sense of interaction created by the interactive behavior. The results highlight potential applications of the synchronization concept both in motor coordination among robotic agents and in enhanced social interaction between humanoid agents and humans.     

In vivo measurement of shoulder joint loads during activities of daily living

Until recently the contact loads acting in the glenohumeral joint have been calculated using musculoskeletal models or measured in vitro. Now, contact forces and moments are measured in vivo using telemeterized shoulder implants. Mean total contact forces from four patients during eight activities of daily living are reported here.Lifting a coffee pot (1.5 kg) with straight arm caused an average force of 105.0%BW (%body weight) (range: 90–124.6%BW), while setting down the coffee pot in the same position led to higher forces of 122.9%BW on the average (105.3–153.4%BW). The highest joint contact forces were measured when the straight arm was abducted or elevated by 90° or more, with a weight in the hand. Lifting up 2 kg from a board up to head height caused a contact force of 98.3%BW (93–103.6%BW); again, setting it down on the board led to higher forces of 131.5%BW (118.8–144.1%BW). In contrast to previously calculated high loads, the contact force during passive holding of a 10 kg weight laterally was only 12.3%BW (9.2–17.9%BW), but when lifting it up to belt height it increased to 91.5%BW (87–95%BW).The moments transferred inside the joint at our patients varied much more than did the forces both inter and intra-individually.Our data suggest that patients with shoulder problems or during the first post-operative weeks after shoulder fractures or joint replacements should avoid certain activities encountered during daily living e.g. lifting or holding a weight with an outstretched arm. Some energy-related optimization criteria used in the literature for analytical musculoskeletal shoulder models must now be reconsidered.     

A model of force and impedance in human arm movements

This paper describes a simple computational model of joint torque and impedance in human arm movements that can be used to simulate three-dimensional movements of the (redundant) arm or leg and to design the control of robots and human-machine interfaces. This model, based on recent physiological findings, assumes that (1) the central nervous system learns the force and impedance to perform a task successfully in a given stable or unstable dynamic environment and (2) stiffness is linearly related to the magnitude of the joint torque and increased to compensate for environment instability. Comparison with existing data shows that this simple model is able to predict impedance geometry well.