Biorobotics/Biomechanics

Spinal cord

Modelling

Neuron

Brain

Analytics

Intracellular

Cerebellum

For three decades our founders challenged the common consensus of one of the biggest scientific questions of our time: How does the brain work?

IntuiCell | Research

Born in the intersection
of AI and Neuroscience.

IntuiCell is a Lund University spin-off with presence in Lund and Stockholm. 

For three decades our founders challenged the common consensus 
of one of the biggest scientific questions of our time:

Today, we apply those groundbreaking neurophysiological insights 
to redesign artificial intelligence. Creating AI that is 

Explore our contrarian research on the mechanisms by which the 
brain predicts the world.

Research

CO Norrsken House
Birger Jarlsgatan 57C
SE-113 56 Stockholm

About

AI isn't actually intelligent. IntuiCell already is.

IntuiCell

How biological brains can learn so quickly to become operational and achieve complex behavior remains an unresolved issue. Here we introduce a ‘neuromorphic’ learning strategy that resembles how immature biological brains learn by consisting of continual random activations of a complex mechanically coupled system with rich, dynamic intrinsic sensor dependencies, in this regard reminiscent of a biological body. Using a dynamic model of biological skin tissue with embedded sensors, we trained small, recurrent networks that emulated a primordial cortex and featured excitatory and inhibitory neurons with simultaneous independent learning in both types of synapses. Training with non-repetitive, random activations of the skin, where the recurrent network activity state was inherited between activations, autonomously led to rapid acquisition of remarkably generalizable representations of a predictive nature. The network could separate inputs and solve a kinematics task that had never been encountered, also after substantial parts of the sensor population were deleted. This strategy of focussing learning on the dominant regularities in dynamic sensory information can potentially explain complex brain operation and efficient learning.

How biological brains can learn so quickly to become operational and achieve complex behavior remains an unresolved issue.

Sensory dependencies rapidly and autonomously yield generalizable representations in recurrent cortical-like networks

Currently, it is accepted that animal locomotion is controlled by a central pattern generator in the spinal cord. Experiments and models show that rhythm generating neurons and genetically determined network properties could sustain oscillatory output activity suitable for locomotion. However, current central pattern generator models do not explain how a spinal cord circuitry, which has the same basic genetic plan across species, can adapt to control the different biomechanical properties and locomotion patterns existing in these species. Here we demonstrate that rhythmic and alternating movements in pendulum models can be learned by a monolayer spinal cord circuitry model using the Bienenstock–Cooper–Munro learning rule, which has been previously proposed to explain learning in the visual cortex. These results provide an alternative theory to central pattern generator models, because rhythm generating neurons and genetically defined connectivity are not required in our model. Though our results are not in contradiction to current models, as existing neural mechanism and structures, not used in our model, can be expected to facilitate the kind of learning demonstrated here. Therefore, our model could be used to augment existing models.

Currently, it is accepted that animal locomotion is controlled by a central pattern generator in the spinal cord.

The Bcm rule allows a spinal cord model to learn rhythmic movements

Time-frequency decomposition is a well-established method to unmix signals generated by multiple sources with unique characteristics. However, there are cases of high signal complexity where existing time-frequency decomposition tools are insufficient for localizing and representing short-bursting signals. One example is the currently highly popular extracellular low-impedance recordings from multi-electrode arrays in the brain 

 where each neuron repeatedly generates a specific signal ‘fingerprint’ (characteristic spike waveform) that can be mixed with the signals of 100s of other sources, including the spikes of nearby neurons. Here we derive the singular superlet transform (SST) method, which enables highly localized representations of fast and short bursts compared to other super-resolution spectral estimators, while also requiring orders of magnitude fewer operations. We demonstrate a substantial edge of SST over current methods in isolating specific neuronal spikes with high-fidelity in challenging, complex recording signals from neocortex 

. We also exemplify SST’s generic signal processing capability by achieving outstanding resolution in the decomposition of complex acoustic data.

Time-frequency decomposition is a well-established method to unmix signals generated by multiple sources with unique characteristics.

Singular superlet transform achieves markedly improved time-frequency super-resolution for separating complex neural signals

The mammalian skin is a densely innervated soft tissue where embedded neural mechanoreceptors report to the brain the mechanical events arising at its surface. Up to now, models of the transformations of these events into neural signals relied on quasistatic or viscoelastic mechanical models of the skin. Here, we employed a model which, in addition to elasticity and viscosity, accounted for mass to accurately reproduce the propagation of mechanical waves observed in vivo. Skin dynamics converted sensory inputs into rapidly evolving spatiotemporal patterns that magnified the information made available to a population of mechanoreceptors. Accounting for dynamics greatly enhanced the separability of tactile inputs and was efficient for a large range of mechanical parameter values. This advantage vanished when these parameters were set to approximate the quasistatic or viscoelastic cases.

The mammalian skin is a densely innervated soft tissue where embedded neural mechanoreceptors report to the brain the mechanical events arising at its surface.

The Import of Skin Tissue Dynamics in Tactile Sensing

The dynamics of the human body can be described by the accelerations and masses of the different body parts (e.g., legs, arm, trunk). These body parts can exhibit specific coordination patterns with each other. In human walking, we found that the swing leg cooperates with the upper body and the stance leg in different ways (e.g., in-phase and out-of-phase in vertical and horizontal directions, respectively). Such patterns of self-assistance found in human locomotion could be of advantage in robotics design, in the design of any assistive device for patients with movement impairments. It can also shed light on several unexplained infrastructural features of the CNS motor control. Self-assistance means that distributed parts of the body contribute to an overlay of functions that are required to solve the underlying motor task. To draw advantage of self-assisting effects, precise and balanced spatiotemporal patterns of muscle activation are necessary. We show that the necessary neural connectivity infrastructure to achieve such muscle control exists in abundance in the spinocerebellar circuitry. We discuss how these connectivity patterns of the spinal interneurons appear to be present already perinatally but also likely are learned. We also discuss the importance of these insights into whole body locomotion for the successful design of future assistive devices and the sense of control that they could ideally confer to the user.

The dynamics of the human body can be described by the accelerations and masses of the different body parts (e.g., legs, arm, trunk).

Whole Body Coordination for Self-Assistance in Locomotion

Biological cortical networks are potentially fully recurrent networks without any distinct output layer, where recognition may instead rely on the distribution of activity across its neurons. Because such biological networks can have rich dynamics, they are well-designed to cope with dynamical interactions of the types that occur in nature, while traditional machine learning networks may struggle to make sense of such data. Here we connected a simple model neuronal network (based on the 'linear summation neuron model' featuring biologically realistic dynamics (LSM), consisting of 10 of excitatory and 10 inhibitory neurons, randomly connected) to a robot finger with multiple types of force sensors when interacting with materials of different levels of compliance. Scope: to explore the performance of the network on classification accuracy. Therefore, we compared the performance of the network output with principal component analysis of statistical features of the sensory data as well as its mechanical properties. Remarkably, even though the LSM was a very small and untrained network, and merely designed to provide rich internal network dynamics while the neuron model itself was highly simplified, we found that the LSM outperformed these other statistical approaches in terms of accuracy.

Biological cortical networks are potentially fully recurrent networks without any distinct output layer, where recognition may instead rely on the distribution of activity across its neurons.

Dynamic neuronal networks efficiently achieve classification in robotic interactions with real-world objects

 that was able to learn and discriminate 

 model and homeostatic synaptic learning mechanism. The first order neuron model was used to emulate biological tactile afferents and the second order neuron model was used to emulate biological cuneate neurons. We have evaluated 10 naturalistic textures using a passive touch protocol, under varying sensing conditions. Tactile sensor data acquired with five textures under five sensing conditions were used for a synaptic learning process, to tune the 

 between tactile afferents and cuneate neurons. Using post-learning synaptic weights, we evaluated the individual and population cuneate neuron responses by decoding across 10 stimuli, under varying sensing conditions. This resulted in a high decoding performance. We further validated the decoding performance across stimuli, irrespective of sensing velocities using a set of 25 cuneate neuron responses. This resulted in a median decoding performance of 96% across the set of cuneate neurons. Being able to learn and perform generalized discrimination across 

, makes this functional spiking tactile system effective and suitable for further 

We implemented a functional neuronal network that was able to learn and discriminate haptic features from biomimetic tactile sensor inputs using a two-layer spiking neuron model and homeostatic synaptic learning mechanism.

Cuneate spiking neural network learning to classify naturalistic texture stimuli under varying sensing conditions

Locomotion control in mammals has been hypothesized to be governed by a central pattern generator (CPG) located in the circuitry of the spinal cord. The most common model of the CPG is the half center model, where two pools of neurons generate alternating, oscillatory activity. In this model, the pools reciprocally inhibit each other ensuring alternating activity. There is experimental support for reciprocal inhibition. However another crucial part of the half center model is a self inhibitory mechanism which prevents the neurons of each individual pool from infinite firing. Self-inhibition is hence necessary to obtain alternating activity. But critical parts of the experimental bases for the proposed mechanisms for self-inhibition were obtained 

, in preparations of juvenile animals. The commonly used adaptation of spike firing does not appear to be present in adult animals 

. We therefore modeled several possible self inhibitory mechanisms for locomotor control. Based on currently published data, previously proposed hypotheses of the self inhibitory mechanism, necessary to support the CPG hypothesis, seems to be put into question by functional evaluation tests or by 

 data. This opens for alternative explanations of how locomotion activity patterns in the adult mammal could be generated.

Locomotion control in mammals has been hypothesized to be governed by a central pattern generator (CPG) located in the circuitry of the spinal cord.

Biological data questions the support of the self inhibition required for pattern generation in the half center model

Sensory feedback from touch is critical for many tasks carried out by robots and humans, such as grasping objects or identifying materials. Electronic skin (e-skin) is a crucial technology for these purposes. Artificial tactile skin that can play the roles of human skin remains a distant possibility because of hard issues in resilience, manufacturing, mechanics, sensorics, electronics, energetics, information processing, and transport. Taken together, these issues make it difficult to bestow robots, or prosthetic devices, with effective tactile skins. Nonetheless, progress over the past few years in relation with the above issues has been encouraging, and we have achieved close to providing some of the abilities of biological skin with the advent of deformable sensors and flexible electronics. The naive imitation of skin morphology and sensing an impoverished set of mechanical and thermal quantities are not sufficient. There is a need to find more efficient ways to extract tactile information from mechanical contact than those previously available. Renewed interest in neuromorphic tactile skin is expected to bring some fresh ideas in this field. This article reviews these new developments, particularly related to the handling of tactile data, energy autonomy, and large-area manufacturing. The challenges in relation with these advances for tactile sensing and haptics in robotics and prosthetics are discussed along with potential solutions.

Sensory feedback from touch is critical for many tasks carried out by robots and humans, such as grasping objects or identifying materials.

Large-area soft e-skin: The challenges beyond sensor designs

Generalization ability in tactile sensing for robotic manipulation is a prerequisite to effectively perform tasks in ever-changing environments. In particular, performing dynamic tactile perception is currently beyond the ability of robotic devices. A biomimetic approach to achieve this dexterity is to develop machines combining compliant robotic manipulators with neuroinspired architectures displaying computational adaptation. Here we demonstrate the feasibility of this approach for dynamic touch tasks experimented by integrating our sensing apparatus in a 6 degrees of freedom robotic arm via a soft wrist. We embodied in the system a model of spike-based neuromorphic encoding of tactile stimuli, emulating the discrimination properties of cuneate nucleus neurons based on pathways with differential delay lines. These strategies allowed the system to correctly perform a dynamic touch protocol of edge orientation recognition (ridges from 0 to 40°, with a step of 5°). Crucially, the task was robust to contact noise and was performed with high performance irrespectively of sensing conditions (sensing forces and velocities). These results are a step forward toward the development of robotic arms able to physically interact in real-world environments with tactile sensing.

Generalization ability in tactile sensing for robotic manipulation is a prerequisite to effectively perform tasks in ever-changing environments.

Born in the intersection
of AI and Neuroscience.

IntuiCell is a Lund University spin-off with presence in Lund and Stockholm.

For three decades our founders challenged the common consensus
of one of the biggest scientific questions of our time:

How does the brain work?

Today, we apply those groundbreaking neurophysiological insights
to redesign artificial intelligence. Creating AI that is actually intelligent.

Explore our contrarian research on the mechanisms by which the
brain predicts the world.

All Publications.

Born in the intersectionof AI and Neuroscience.

IntuiCell is a Lund University spin-off with presence in Lund and Stockholm.

For three decades our founders challenged the common consensus of one of the biggest scientific questions of our time:

How does the brain work?

Today, we apply those groundbreaking neurophysiological insights to redesign artificial intelligence. Creating AI that is actually intelligent.

Explore our contrarian research on the mechanisms by which the brain predicts the world.

All Publications.

Born in the intersection
of AI and Neuroscience.

For three decades our founders challenged the common consensus
of one of the biggest scientific questions of our time:

Today, we apply those groundbreaking neurophysiological insights
to redesign artificial intelligence. Creating AI that is actually intelligent.

Explore our contrarian research on the mechanisms by which the
brain predicts the world.