Biomimetic vibrissal sensing for robots

Active vibrissal touch can be used to replace or to supplement sensory systems such as computer vision and, therefore, improve the sensory capacity of mobile robots. This paper describes how arrays of whisker-like touch sensors have been incorporated onto mobile robot platforms taking inspiration from biology for their morphology and control. There were two motivations for this work: first, to build a physical platform on which to model, and therefore test, recent neuroethological hypotheses about vibrissal touch; second, to exploit the control strategies and morphology observed in the biological analogue to maximize the quality and quantity of tactile sensory information derived from the artificial whisker array. We describe the design of a new whiskered robot, Shrewbot, endowed with a biomimetic array of individually controlled whiskers and a neuroethologically inspired whisking pattern generation mechanism. We then present results showing how the morphology of the whisker array shapes the sensory surface surrounding the robot's head, and demonstrate the impact of active touch control on the sensory information that can be acquired by the robot. We show that adopting bio-inspired, low latency motor control of the rhythmic motion of the whiskers in response to contact-induced stimuli usefully constrains the sensory range, while also maximizing the number of whisker contacts. The robot experiments also demonstrate that the sensory consequences of active touch control can be usefully investigated in biomimetic robots.     

Fast Feedback in Active Sensing: Touch-Induced Changes to Whisker-Object Interaction

Whisking mediated touch is an active sense whereby whisker movements are modulated by sensory input and behavioral context. Here we studied the effects of touching an object on whisking in head-fixed rats. Simultaneous movements of whiskers C1, C2, and D1 were tracked bilaterally and their movements compared. During free-air whisking, whisker protractions were typically characterized by a single acceleration-deceleration event, whisking amplitude and velocity were correlated, and whisk duration correlated with neither amplitude nor velocity. Upon contact with an object, a second acceleration-deceleration event occurred in about 25% of whisk cycles, involving both contacting (C2) and non-contacting (C1, D1) whiskers ipsilateral to the object. In these cases, the rostral whisker (C2) remained in contact with the object throughout the double-peak phase, which effectively prolonged the duration of C2 contact. These “touch-induced pumps” (TIPs) were detected, on average, 17.9 ms after contact. On a slower time scale, starting at the cycle following first touch, contralateral amplitude increased while ipsilateral amplitude decreased. Our results demonstrate that sensory-induced motor modulations occur at various timescales, and directly affect object palpation.     

Active vibrissal sensing in rodents and marsupials

In rats, the long facial whiskers (mystacial macrovibrissae) are repetitively and rapidly swept back and forth during exploration in a behaviour known as 'whisking'. In this paper, we summarize previous evidence from rats, and present new data for rat, mouse and the marsupial grey short-tailed opossum (Monodelphis domestica) showing that whisking in all three species is actively controlled both with respect to movement of the animal's body and relative to environmental structure. Using automatic whisker tracking, and Fourier analysis, we first show that the whisking motion of the mystacial vibrissae, in the horizontal plane, can be approximated as a blend of two sinusoids at the fundamental frequency (mean 8.5, 11.3 and 7.3 Hz in rat, mouse and opossum, respectively) and its second harmonic. The oscillation at the second harmonic is particularly strong in mouse (around 22 Hz) consistent with previous reports of fast whisking in that species. In all three species, we found evidence of asymmetric whisking during head turning and following unilateral object contacts consistent with active control of whisker movement. We propose that the presence of active vibrissal touch in both rodents and marsupials suggests that this behavioural capacity emerged at an early stage in the evolution of therian mammals.     

Electromyographic activity of mystacial pad musculature during whisking behavior in the rat

Cinematographic measurements of whisker movements generated by behaving rats were compared with electromyographic (EMG) activity recorded simultaneously from mystacial pad musculature. Muscle activity consisted of repetitive bursts, each of which initiated a "whisking" cycle consisting of a protraction followed by a retraction. Protraction amplitude and velocity were directly proportional to the amount of EMG activity during forward whisker movement. Overtime, the intensity of muscle discharge determined the set point about which the vibrissae moved; higher levels of muscle activity resulted in a greater degree of overall whisker protraction. These findings are consistent with the known anatomy of the facial musculature and underscore the importance of whisker protraction in the acquisition of tactile information by the vibrissae.     

"Real-time" monitoring of vibrissa contacts during rodent whisking

Rodent whisking behavior provides active touch as input into a widely studied model system of information processing and behavior. We previously developed a simple optoelectronic system to monitor whisker movements in "real time" in head held rats at rest or performing various tasks such as tactile discrimination. We now describe a simple piezioelectic film device for detecting initial whisker contacts during whisking also in real time. In some applications this is as effective as high-speed videos and can be configured to isolate the contacts from different whiskers. The construction of this simple device is detailed. In addition to providing information during recordings from awake animals, the device could be used, for example, as an operant "manipulandum" for contingent reinforcement of object detection with a whisker.     

Morphogenesis and renewal of hair follicles from adult multipotent stem cells

The upper region of the outer root sheath of vibrissal follicles of adult mice contains multipotent stem cells that respond to morphogenetic signals to generate multiple hair follicles, sebaceous glands, and epidermis, i.e., all the lineages of the hairy skin. At the time when hair production ceases and when the lower region of the follicle undergoes major structural changes, the lower region contains a significant number of clonogenic keratinocytes, and can then respond to morphogenetic signals. This demonstrates that multipotent stem cells migrate to the root of the follicle to produce whisker growth. Moreover, our results indicate that the clonogenic keratinocytes are closely related, if not identical, to the multipotent stem cells, and that the regulation of whisker growth necessitates a precise control of stem cell trafficking.     

“Real-time” monitoring of vibrissa contacts during rodent whisking

Rodent whisking behavior provides active touch as input into a widely studied model system of information processing and behavior. We previously developed a simple optoelectronic system to monitor whisker movements in “real time” in head held rats at rest or performing various tasks such as tactile discrimination. We now describe a simple piezioelectic film device for detecting initial whisker contacts during whisking also in real time. In some applications this is as effective as high-speed videos and can be configured to isolate the contacts from different whiskers. The construction of this simple device is detailed. In addition to providing information during recordings from awake animals, the device could be used, for example, as an operant manipulandum for contingent reinforcement of object detection with a whisker.     

Discriminative whisking in the head-fixed rat: optoelectronic monitoring during tactile detection and discrimination tasks

We compared whisking movement patterns during acquisition of tactile detection and object discrimination under conditions in which (a) head movements are excluded and (b) exposure to tactile discriminanda is confined to the large, moveable vibrissae (macrovibrissae). We used optoelectronic instrumentation to track the movements of an individual whisker with high spatio-temporal resolution and a testing paradigm, which allowed us to dissociate performance on an "indicator" response (lever pressing) from the rat's "observing" responses (discriminative whisking). We analyzed the relation between discrimination performance and whisking movement patterns in order to clarify the process by which the indicator response comes under the stimulus control of information acquired by the rat's whisking behavior. Whisking patterns over the course of task acquisition differed with task demands. Acquisition of the Detection task was correlated with modulation of only one whisking movement parameter-total number of whisks emitted, and more whisking was seen on trials in which the discriminandum was absent. Discrimination between a sphere and cube differing in size and texture was correlated with a reduction in whisk duration and protraction amplitude and with a shift towards higher whisking frequencies. Our findings confirm previous reports that acquisition of tactile discriminations involves modulation by the animal of both the amount and the type of whisking. In contrast with a previous report (Brecht et al., 1997), they indicate that rats can solve tactile object detection and discrimination tasks (a) using only the large, motile mystacial vibrissae (macrovibrissae) and (b) without engaging in head movements. We conclude that the functional contribution of the macrovibrissae will vary with the nature of the task and the conditions of testing.     

An internal model architecture for novelty detection: implications for cerebellar and collicular roles in sensory processing

The cerebellum is thought to implement internal models for sensory prediction, but details of the underlying circuitry are currently obscure. We therefore investigated a specific example of internal-model based sensory prediction, namely detection of whisker contacts during whisking. Inputs from the vibrissae in rats can be affected by signals generated by whisker movement, a phenomenon also observable in whisking robots. Robot novelty-detection can be improved by adaptive noise-cancellation, in which an adaptive filter learns a forward model of the whisker plant that allows the sensory effects of whisking to be predicted and thus subtracted from the noisy sensory input. However, the forward model only uses information from an efference copy of the whisking commands. Here we show that the addition of sensory information from the whiskers allows the adaptive filter to learn a more complex internal model that performs more robustly than the forward model, particularly when the whisking-induced interference has a periodic structure. We then propose a neural equivalent of the circuitry required for adaptive novelty-detection in the robot, in which the role of the adaptive filter is carried out by the cerebellum, with the comparison of its output (an estimate of the self-induced interference) and the original vibrissal signal occurring in the superior colliculus, a structure noted for its central role in novelty detection. This proposal makes a specific prediction concerning the whisker-related functions of a region in cerebellar cortical zone A(2) that in rats receives climbing fibre input from the superior colliculus (via the inferior olive). This region has not been observed in non-whisking animals such as cats and primates, and its functional role in vibrissal processing has hitherto remained mysterious. Further investigation of this system may throw light on how cerebellar-based internal models could be used in broader sensory, motor and cognitive contexts.     

Discriminative whisking in the head-fixed rat: optoelectronic monitoring during tactile detection and discrimination tasks

We compared whisking movement patterns during acquisition of tactile detection and object discrimination under conditions in which (a) head movements are excluded and (b) exposure to tactile discriminanda is confined to the large, moveable vibrissae (macrovibrissae). We used optoelectronic instrumentation to track the movements of an individual whisker with high spatio-temporal resolution and a testing paradigm, which allowed us to dissociate performance on an “indicator” response (lever pressing) from the rat's “observing” responses (discriminative whisking). We analyzed the relation between discrimination performance and whisking movement patterns in order to clarify the process by which the indicator response comes under the stimulus control of information acquired by the rat's whisking behavior. Whisking patterns over the course of task acquisition differed with task demands. Acquisition of the Detection task was correlated with modulation of only one whisking movement parameter - total number of whisks emitted, and more whisking was seen on trials in which the discriminandum was absent. Discrimination between a sphere and cube differing in size and texture was correlated with a reduction in whisk duration and protraction amplitude and with a shift towards higher whisking frequencies. Our findings confirm previous reports that acquisition of tactile discriminations involves modulation by the animal of both the amount and the type of whisking. In contrast with a previous report (Brecht et al., 1997), they indicate that rats can solve tactile object detection and discrimination tasks (a) using only the large, motile mystacial vibrissae (macrovibrissae) and (b) without engaging in head movements.We conclude that the functional contribution of the macrovibrissae will vary with the nature of the task and the conditions of testing.     

The Advantages of a Tapered Whisker

The role of facial vibrissae (whiskers) in the behavior of terrestrial mammals is principally as a supplement or substitute for short-distance vision. Each whisker in the array functions as a mechanical transducer, conveying forces applied along the shaft to mechanoreceptors in the follicle at the whisker base. Subsequent processing of mechanoreceptor output in the trigeminal nucleus and somatosensory cortex allows high accuracy discriminations of object distance, direction, and surface texture. The whiskers of terrestrial mammals are tapered and approximately circular in cross section. We characterize the taper of whiskers in nine mammal species, measure the mechanical deflection of isolated felid whiskers, and discuss the mechanics of a single whisker under static and oscillatory deflections. We argue that a tapered whisker provides some advantages for tactile perception (as compared to a hypothetical untapered whisker), and that this may explain why the taper has been preserved during the evolution of terrestrial mammals.     

Vibrissal kinematics in 3D: tight coupling of azimuth, elevation, and torsion across different whisking modes

Perception is usually an active process by which action selects and affects sensory information. During rodent active touch, whisker kinematics influences how objects activate sensory receptors. In order to fully characterize whisker motion, we reconstructed whisker position in 3D and decomposed whisker motion to all its degrees of freedom. We found that, across behavioral modes, in both head-fixed and freely moving rats, whisker motion is characterized by translational movements and three rotary components: azimuth, elevation, and torsion. Whisker torsion, which has not previously been described, was large (up to 100 degrees), and torsional angles were highly correlated with whisker azimuths. The coupling of azimuth and torsion was consistent across whisking epochs and rats and was similar along rows but systematically varied across rows such that rows A and E counterrotated. Torsional rotation of the whiskers enables contact information to be mapped onto the circumference of the whisker follicles in a predictable manner across protraction-retraction cycles.     

Feedback control in active sensing: rat exploratory whisking is modulated by environmental contact

Rats sweep their facial whiskers back and forth to generate tactile sensory information through contact with environmental structure. The neural processes operating on the signals arising from these whisker contacts are widely studied as a model of sensing in general, even though detailed knowledge of the natural circumstances under which such signals are generated is lacking. We used digital video tracking and wireless recording of mystacial electromyogram signals to assess the effects of whisker-object contact on whisking in freely moving animals exploring simple environments. Our results show that contact leads to reduced protraction (forward whisker motion) on the side of the animal ipsilateral to an obstruction and increased protraction on the contralateral side. Reduced ipsilateral protraction occurs rapidly and in the same whisk cycle as the initial contact. We conclude that whisker movements are actively controlled so as to increase the likelihood of environmental contacts while constraining such interactions to involve a gentle touch. That whisking pattern generation is under strong feedback control has important implications for understanding the nature of the signals reaching upstream neural processes.     

Facial morphology and vibrissal movement in the golden hamster

The major cranial vibrissae in the golden hamster can be moved in complex ways that suggest they are served by a finely controlled motor system. Movements are hypothesized to be the products of differential blood flow and pressure regulation in the sinus surrounding each vibrissal follicle, contractions of the striated facial muscles, and elastic rebound in the connective tissues. The vasculature contributes hydrostatic forces that erect the vibrissae slightly and distort their connective tissue bedding, rigidify the vibrissal capsules, thus forming firm bases of attachment for certain facial muscles, and theoretically provide a pressure plate around the follicle, important in lowering the firing thresholds of receptor endings. The facial muscles supply the major forces in erection and protraction of the vibrissae by acting on both the capsules and the connective tissue bedding. The connective tissues are organized into capsular and extracapsular systems that serve to stabilize the vibrissae and return them to initial rest positions. The slight movements of the genal vibrissa are the effects of vascular and connective tissue dynamics, the musculature being uninvolved. Wide angle movements of the supraorbital vibrissae are products of the vasculature and connective tissues, plus contractions of the Mm. orbicularis oculi and frontalis. Mystacial vibrissal movement is quite complex. The vasculature supplies a small degree of capsular erection and mystacial pad distortion, but primarily rigidifies the capsules. The bulk of erection and protraction is produced by the M. nasolabialis profundus (NLP) and the vibrissal capsular muscles (VCM). The NLP distorts the mystacial pad; the VCM tilt the capsules relative to the pad. Retraction is mainly accomplished by elastic rebound in the pad, this being aided in its extreme degrees by the Mm. nasolabialis and maxillolabialis. The Mm. nasolabialis superficialis and buccinator pars orbicularis oris help to spread the vibrissae into a dorsoventral fan and stabilize the mystacial pad during whisking.     

Extensive Hair-Shaft Elongation by Isolated Mouse Whisker Follicles in Very Long-Term Gelfoam? Histoculture

We have previously studied mouse whisker follicles in Gelfoam® histoculture to determine the role of nestin-expressing plutipotent stem cells, located within the follicle, in the growth of the follicular sensory nerve. Long-term Gelfoam® whisker histoculture enabled hair follicle nestin-expressing stem cells to promote the extensive elongation of the whisker sensory nerve, which contained axon fibers. Transgenic mice in which the nestin promoter drives green fluorescent protein (ND-GFP) were used as the source of the whiskers allowing imaging of the nestin-expressing stem cells as they formed the follicular sensory nerve. In the present report, we show that Gelfoam®-histocultured whisker follicles produced growing pigmented and unpigmented hair shafts. Hair-shaft length increased rapidly by day-4 and continued growing until at least day-12 after which the hair-shaft length was constant. By day-63 in histoculture, the number of ND-GFP hair follicle stem cells increased significantly and the follicles were intact. The present study shows that Gelfoam® histoculture can support extensive hair-shaft growth as well as hair follicle sensory-nerve growth from isolated hair follicles which were maintained over very long periods of time. Gelfoam® histoculture of hair follicles can provide a very long-term period for evaluating novel agents to promote hair growth.     

Structure of the sinus hair follicle in the big-clawed shrew,Sorex unguiculatus

The structural features of sinus hair follicles in Sorex unguiculatus were studied by macroscopic dissection, serial section light microscopy and electron microscopy. The shrew has about 540 sinus hairs regularly arranged on the snout. The maxillary nerves innervating them are extremely thick, while the optic nerves are very thin. Thus the follicle must be one of the most important sense organs in this animal. In the follicle the ring sinus is well-developed and the trabeculae of the cavernous sinus are reduced in number and thickness. The ring bulge is not a unified structure but a pair of bodies which consist of head, stalk and attachment plaque. It is characterized by the presence of numberous thick collagen fibrils (400 nm) and appears to be mechanically rigid. Lanceolate nerve terminals, free endings, Merkel cells with nerve terminals and unmyelinated fibers are observed, but encapsulated endings are lacking in and around the follicles. Straight lanceolate terminals on the posterior side of the follicle are thick and three-sided in cross section, while those on the anterior side are thin and two-sided. Free endings are located on the anterior side of the follicle. These and other findings are discussed on the basis of the assumption that the Sorex sinus hair follicle is more specialized as a vibrating system than in other mammals.     

Development of Merkel cell populations with contrasting sensitivities to neonatal deafferentation in the rat whisker pad

In this study, we used the quinacrine fluorescence technique to investigate the embryonic and early postnatal development of two distinct populations of Merkel cells in the rat whisker pad and the consequences of neonatal deafferentation on their subsequent development. Annular clusters of Merkel cells first appear in the epidermis near the caudal margin of the mystacial region between embryonic days E14 and E15 at dome sites located on horizontal ridges where the primordial vibrissal follicles develop. The development of these cells progresses in a caudorostral sequence across the whisker pad as does the development of the vibrissal follicles. Each cluster eventually forms a conical ridge or collar of about 130 Merkel cells that surrounds the vibrissal hair shaft as it penetrates the overlying pad epidermis. In the vibrissae, which develop as downgrowths from the horizontal ridges at the dome sites, Merkel cells first appear (caudally) between E16 and E17 and form a cylindrical cuff within the outer root sheath; cells are added progressively until about the end of the first postnatal week when a plateau level of about 750-800 cells is reached. Following unilateral transection of the infraorbital nerve at 24-36 hr after birth, these vibrissal Merkel cells continued to develop along a time course that was indistinguishable from normal, at least over the first 2 weeks of postnatal life. In contrast, all or most of the Merkel cells that normally develop within collars or annular clusters in the pad epidermis (around both the vibrissal and intervibrissal or pelage hairs) either disappeared within a few days or failed to develop. Other light and electron microscopic procedures supported the main findings and confirmed that the denervation was successful. Thus, the vibrissal Merkel cells, like those in the glabrous hindpaw, behaved as a distinct class which develops postnatally and is maintained (at least over a 2-week period) without the presence of sensory nerves. Since both the mystacial vibrissae and glabrous hindpaw have specialized cortical representations, a possible relationship between these findings and the organization of the somatosensory cortex during development is discussed.     

Arrangement and structure of sinus hair muscles in the big-clawed shrew,Sorex unguiculatus

The arrangement and structure of sinus hair muscles in the snout of the shrew, Sorex unguiculatus, were studied by electron microscopy and serial section light microscopy. Both striated and smooth muscles are directly associated with sinus hair follicles. The striated muscle fibers originate from the base of a follicle and insert onto the superficial portion of adjoining caudally positioned follicles. Some fibers insert into the corium instead of inserting into a follicle. The fibers show a fine structure typical of red fibers. Smooth muscle cells form a network with elastic fibers beneath the corium. Some cells are directly attached to the capsule of the sinus, thus forming a type of M. arrector pili. Striated muscle fibers that appear to end in the corium are connected with the smooth muscle network through the elastic fibers which appear to function as the tendon of these two types of muscle cell.     

Topography of rodent whisking--I. Two-dimensional monitoring of whisker movements

During 'active touch' the rodent whiskers scan the environment in a series of repetitive movements ('whisks') generating afferent signals which transform the spatial properties of objects into spatio-temporal patterns of neural activity. Based upon analyses carried out in a single movement plane, it has been generally assumed that these trajectories are essentially uni-dimensional, although more complex movements have been described in some rodents. The present study was designed to examine this assumption and to more precisely characterize whisking topography by monitoring whisking trajectories along both the antero-posterior and dorso-ventral axes. Using optoelectronic monitoring techniques with high-spatio-temporal resolution, movement data were obtained from a population of vibrissae sampled at different locations on the mystacial pad in head-fixed rats isolated from the perturbing effects of contact. For a substantial proportion of the population of whisking movements sampled, the trajectories generated by a single whisker is most accurately described as occupying an expended two-dimensional space in which the A-P component predominates. However, the whisker system exhibits a considerable range of trajectory types, suggesting a high degree of movement flexibility. For each vibrissa position, it was possible to delineate a 'trajectory' domain -- that portion of the animal's whisking space which is scanned by the movements of that vibrissa during whisking. Since the 'domains' of adjacent whiskers in the same row tend to overlap, synchronized movements of a subset of whiskers could generate a set of overlapping somatosensory fields analogous to overlapping retinal receptive fields. The organization of such trajectory domains within the rats' whisking space could provide the spatial component of the spatio-temporal integration process required to extract information about environmental features from the inputs generated by its recursive whisking movements.     

Long-Term Extensive Ectopic Hair Growth on the Spinal Cord of Mice from Transplanted Whisker Follicles

We have previously demonstrated that hair follicles contain nestin-expressing pluripotent stem cells that can effect nerve and spinal cord repair upon transplantation. In the present study, isolated whisker follicles from nestin-driven green fluorescent protein (ND-GFP) mice were histocultured on Gelfoam for 3 weeks for the purpose of transplantation to the spinal cord to heal an induced injury. The hair shaft was cut off from Gelfoam-histocultured whisker follicles, and the remaining part of the whisker follicles containing GFP-nestin expressing pluripotent stem cells were transplanted into the injured spinal cord of nude mice, along with the Gelfoam. After 90 days, the mice were sacrificed and the spinal cord lesion was observed to have healed. ND-GFP expression was intense at the healed area of the spinal cord, as observed by fluorescence microscopy, demonstrating that the hair follicle stem cells were involved in healing the spinal cord. Unexpectedly, the transplanted whisker follicles sprouted out remarkably long hair shafts in the spinal cord during the 90 days after transplantation of Gelfoam whisker histocultures to the injured spine. The pigmented hair fibers, grown from the transplanted whisker histocultures, curved and enclosed the spinal cord. The unanticipated results demonstrate the great potential of hair growth after transplantation of Gelfoam hair follicle histocultures, even at an ectopic site.