Bio-inspired design strategies for central pattern generator control in modular robotics

New findings in the nervous system of invertebrates have shown how a number of features of central pattern generator (CPG) circuits contribute to the generation of robust flexible rhythms. In this paper we consider recently revealed strategies that living CPGs follow to design CPG control paradigms for modular robots. To illustrate them, we divide the task of designing an example CPG for a modular robot into independent problems. We formulate each problem in a general way and provide a bio-inspired solution for each of them: locomotion information coding, individual module control and inter-module coordination. We analyse the stability of the CPG numerically, and then test it on a real robot. We analyse steady state locomotion and recovery after perturbations. In both cases, the robot is able to autonomously find a stable effective locomotion state. Finally, we discuss how these strategies can result in a more general design approach for CPG-based locomotion.     

Application of the mathematics of coupled oscillator systems to the analysis of the neural control of locomotion.

In many animals, the activities of limb motor neurons are rhythmic during locomotion. In some animals it is known that each limb is innervated by a local control center that resides in a discrete portion of the central nervous system. Each local control center is a biological oscillator. Since each limb moves with the same frequency as each other limb and with regulated phase delay with respect to each other limb, then it follows that the local control centers are coupled to one another. The locomotory pattern generator within the central nervous system is therefore a coupled oscillator system. The mathematics of coupled oscillator systems can assist in the construction of a model of the neural pattern generator. This model can be utilized to formulate testable predictions concerning the neural control of locomotion. Experimental data gathered from organisms in several phylums are consistent with the predictions of the model.     

Mantisbot is a robotic model of visually guided motion in the praying mantis

Insects use highly distributed nervous systems to process exteroception from head sensors, compare that information with state-based goals, and direct posture or locomotion toward those goals. To study how descending commands from brain centers produce coordinated, goal-directed motion in distributed nervous systems, we have constructed a conductance-based neural system for our robot MantisBot, a 29 degree-of-freedom, 13.3:1 scale praying mantis robot. Using the literature on mantis prey tracking and insect locomotion, we designed a hierarchical, distributed neural controller that establishes the goal, coordinates different joints, and executes prey-tracking motion. In our controller, brain networks perceive the location of prey and predict its future location, store this location in memory, and formulate descending commands for ballistic saccades like those seen in the animal. The descending commands are simple, indicating only 1) whether the robot should walk or stand still, and 2) the intended direction of motion. Each joint's controller uses the descending commands differently to alter sensory-motor interactions, changing the sensory pathways that coordinate the joints' central pattern generators into one cohesive motion. Experiments with one leg of MantisBot show that visual input produces simple descending commands that alter walking kinematics, change the walking direction in a predictable manner, enact reflex reversals when necessary, and can control both static posture and locomotion with the same network.     

Visually guided gait modifications for stepping over an obstacle: a bio-inspired approach

There is an increasing interest in conceiving robotic systems that are able to move and act in an unstructured and not predefined environment, for which autonomy and adaptability are crucial features. In nature, animals are autonomous biological systems, which often serve as bio-inspiration models, not only for their physical and mechanical properties, but also their control structures that enable adaptability and autonomy—for which learning is (at least) partially responsible. This work proposes a system which seeks to enable a quadruped robot to online learn to detect and to avoid stumbling on an obstacle in its path. The detection relies in a forward internal model that estimates the robot’s perceptive information by exploring the locomotion repetitive nature. The system adapts the locomotion in order to place the robot optimally before attempting to step over the obstacle, avoiding any stumbling. Locomotion adaptation is achieved by changing control parameters of a central pattern generator (CPG)-based locomotion controller. The mechanism learns the necessary alterations to the stride length in order to adapt the locomotion by changing the required CPG parameter. Both learning tasks occur online and together define a sensorimotor map, which enables the robot to learn to step over the obstacle in its path. Simulation results show the feasibility of the proposed approach.     

Multi-layered multi-pattern CPG for adaptive locomotion of humanoid robots

In this paper, we present an extended mathematical model of the central pattern generator (CPG) in the spinal cord. The proposed CPG model is used as the underlying low-level controller of a humanoid robot to generate various walking patterns. Such biological mechanisms have been demonstrated to be robust in locomotion of animal. Our model is supported by two neurophysiological studies. The first study identified a neural circuitry consisting of a two-layered CPG, in which pattern formation and rhythm generation are produced at different levels. The second study focused on a specific neural model that can generate different patterns, including oscillation. This neural model was employed in the pattern generation layer of our CPG, which enables it to produce different motion patterns—rhythmic as well as non-rhythmic motions. Due to the pattern-formation layer, the CPG is able to produce behaviors related to the dominating rhythm (extension/flexion) and rhythm deletion without rhythm resetting. The proposed multi-layered multi-pattern CPG model (MLMP-CPG) has been deployed in a 3D humanoid robot (NAO) while it performs locomotion tasks. The effectiveness of our model is demonstrated in simulations and through experimental results.     

A proprioceptive neuromechanical theory of crawling

The locomotion of many soft-bodied animals is driven by the propagation of rhythmic waves of contraction and extension along the body. These waves are classically attributed to globally synchronized periodic patterns in the nervous system embodied in a central pattern generator (CPG). However, in many primitive organisms such as earthworms and insect larvae, the evidence for a CPG is weak, or even non-existent. We propose a neuromechanical model for rhythmically coordinated crawling that obviates the need for a CPG, by locally coupling the local neuro-muscular dynamics in the body to the mechanics of the body as it interacts frictionally with the substrate. We analyse our model using a combination of analytical and numerical methods to determine the parameter regimes where coordinated crawling is possible and compare our results with experimental data. Our theory naturally suggests mechanisms for how these movements might arise in developing organisms and how they are maintained in adults, and also suggests a robust design principle for engineered motility in soft systems.     

Stress-induced thermotolerance of ventilatory motor pattern generation in the locust, Locusta migratoria

Ventilation is a crucial motor activity that provides organisms with an adequate circulation of respiratory gases. For animals that exist in harsh environments, an important goal is to protect ventilation under extreme conditions. Heat shock, anoxia, and cold shock are environmental stresses that have previously been shown to trigger protective responses. We used the locust to examine stress-induced thermotolerance by monitoring the ability of the central nervous system to generate ventilatory motor patterns during a subsequent heat exposure. Preparations from pre-stressed animals had an increased incidence of motor pattern recovery following heat-induced failure, however, prior stress did not alter the characteristics of the ventilatory motor pattern. During constant heat exposure at sub-lethal temperatures, we observed a protective effect of heat shock pre-treatment. Serotonin application had similar effects on motor patterns when compared to prior heat shock. These studies are consistent with previous studies that indicate prior exposure to extreme temperatures and hypoxia can protect neural operation against high temperature stress. They further suggest that the protective mechanism is a time-dependent process best revealed during prolonged exposure to extreme temperatures and is mediated by a neuromodulator such as serotonin.     

Development of an Amphibious Turtle-Inspired Spherical Mother Robot

Robots play an important role in underwater monitoring and recovery operations, such as pollution detection, submarine sampling and data collection, video mapping, and object recovery in dangerous places. However, regular-sized robots may not be suitable for applications in some restricted underwater environments. Accordingly, in previous research we designed several novel types of bio-inspired microrobots using Ionic Polymer Metal Composite （IPMC） and Shape Memory Alloy （SMA） ac- tuators. These microrobots possess some attributes of compact structure, multi-functionality, flexibility, and precise positioning. However, they lack the attributes of long endurance, stable high speed, and large load capacity necessary for real-world appli- cations. To overcome these disadvantages, we proposed a mother-son robot system, composed of several microrobots as sons and a newly designed amphibious spherical robot as the mother. Inspired by amphibious turtles, the mother robot was designed with a spherical body and four legs with two Degrees of Freedom （DOF）. It is actuated by four vectored water-jet propellers and ten servomotors, and it is capable of walking on land and cruising underwater. We analysed the mother robot＇s walking and underwater cruising mechanisms, constructed a prototype, and carried out a series of experiments to evaluate its amphibious motions. Good motion performance was observed in the experiments.     

Modulation of swimming speed in the pteropod mollusc,Clione limacina: Role of a compartmental serotonergic system

In locomotory systems, the central pattern generator and motoneuron output must be modulated in order to achieve variability in locomotory speed, particularly when speed changes are important components of different behavior acts. The swimming system of the pteropod molluscClione limacina is an excellent model system for investigating such modulation. In particular, a system of central serotonergic neurons has been shown to be intimately involved in regulating output of the locomotory pattern generator and motor system ofClione. There are approximately 27 pairs of serotonin-immunoreactive neurons in the central nervous system ofClione, with about 75% of these identified. The majority of these identified immunoreactive neurons are involved in various aspects of locomotory speed modulation. A symmetrical cluster of pedal serotonergic neurons serves to increase wing contractility without affecting wing-beat frequency or motoneuron activity. Two clusters of cerebral cells produce widespread responses that lead to an increase in pattern generator cycle frequency, recruitment of swim motoneurons, activation of the pedal serotonergic neurons and excitation of the heart excitor neuron. A pair of ventral cerebral neurons provides weak excitatory inputs to the swimming system, and strongly inhibits neurons of the competing whole-body withdrawal network. Overall, the serotonergic system inClione is compartmentalized so that each subsystem (usually neuron cluster) can act independently or in concert to produce variability in locomotory speed.     

From network-to-antibody robustness in a bio-inspired immune system

Behavioural robustness at antibody and immune network level is discussed. The robustness of the immune response that drives an autonomous mobile robot is examined with two computational experiments in the autonomous mobile robots trajectory generation context in unknown environments. The immune response is met based on the immune network metaphor for different low-level behaviours coordination. These behaviours are activated when a robot sense the appropriate conditions in the environment in relation to the network current state. Results are obtained over a case study in computer simulation as well as in laboratory experiments with a Khepera II microrobot. In this work, we develop a set of tests where such an immune response is externally perturbed at network or low-level behavioural modules to analyse the robust capacity of the system to unexpected perturbations. Emergence of robust behaviour and high-level immune response relates to the coupling between behavioural modules that are selectively engaged with the environment based on immune response. Experimental evidence leads discussions on a dynamical systems perspective of behavioural robustness in artificial immune systems that goes beyond the isolated immune network response.     

Experiments on the central pattern generator for swimming in amphibian embryos

The central nervous system of paralysed Xenopus laevis embryos can generate a motor output pattern suitable for swimming locomotion. By recording motor root activity in paralysed embryos with transected nervous systems we have shown that: (a) the spinal cord is capable of swimming pattern generation; (b) swimming pattern generator capability in the hindbrain and spinal cord is distributed; (c) caudal hindbrain is necessary for sustained swimming output after discrete stimulation. By recording similarly from embryos whose central nervous system was divided longitudinally into left and right sides, we have shown that: (a) each side can generate rhythmic motor output with cycle periods like those in swimming; (b) during this activity cycle period increases within an episode, and there is the usual rostrocaudal delay found in swimming; (c) this activity is influenced by sensory stimuli in the same way as swimming activity; (d) normal phase coupling of the left and right sides can be established by the ventral commissure in the spinal cord. We conclude that interactions between the antagonistic (left and right) motor systems are not necessary for swimming rhythm generation and present a model for swimming pattern generation where autonomous rhythm generators on each side of the nervous system drive the motoneurons. Alternation is achieved by reciprocal inhibition, and activity is initiated and maintained by tonic excitation from the hindbrain.     

An analog VLSI implementation of a visual interneuron: enhanced sensory processing through biophysical modeling

Flies are capable of rapid, coordinated flight through unstructured environments. This flight is guided by visual motion information that is extracted from photoreceptors in a robust manner. One feature of the fly's visual processing that adds to this robustness is the saturation of wide-field motion-sensitive neuron responses with increasing pattern size. This makes the cell's responses less dependent on the sparseness of the optical flow field while retaining motion information. By implementing a compartmental neuronal model in silicon, we add this "gain control" to an existing analog VLSI model of fly vision. This results in enhanced performance in a compact, low-power CMOS motion sensor. Our silicon system also demonstrates that modern, biophysically-detailed models of neural sensory processing systems can be instantiated in VLSI hardware.     

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     

Predicting adaptive behavior in the environment from central nervous system dynamics

To generate adaptive behavior, the nervous system is coupled to the environment. The coupling constrains the dynamical properties that the nervous system and the environment must have relative to each other if adaptive behavior is to be produced. In previous computational studies, such constraints have been used to evolve controllers or artificial agents to perform a behavioral task in a given environment. Often, however, we already know the controller, the real nervous system, and its dynamics. Here we propose that the constraints can also be used to solve the inverse problem--to predict from the dynamics of the nervous system the environment to which they are adapted, and so reconstruct the production of the adaptive behavior by the entire coupled system. We illustrate how this can be done in the feeding system of the sea slug Aplysia. At the core of this system is a central pattern generator (CPG) that, with dynamics on both fast and slow time scales, integrates incoming sensory stimuli to produce ingestive and egestive motor programs. We run models embodying these CPG dynamics--in effect, autonomous Aplysia agents--in various feeding environments and analyze the performance of the entire system in a realistic feeding task. We find that the dynamics of the system are tuned for optimal performance in a narrow range of environments that correspond well to those that Aplysia encounter in the wild. In these environments, the slow CPG dynamics implement efficient ingestion of edible seaweed strips with minimal sensory information about them. The fast dynamics then implement a switch to a different behavioral mode in which the system ignores the sensory information completely and follows an internal "goal," emergent from the dynamics, to egest again a strip that proves to be inedible. Key predictions of this reconstruction are confirmed in real feeding animals.     

Neuromodulation of central pattern generators in invertebrates and vertebrates

Central pattern generators are subject to extensive modulation that generates flexibility in the rhythmic outputs of these neural networks. The effects of neuromodulators interact with one another, and modulatory neurons are themselves often subject to modulation, enabling both higher order control and indirect interactions among central pattern generators. In addition, modulators often directly mediate the interactions between functionally related central pattern generators. In systems such as the vertebrate respiratory central pattern generator, multiple pacemaker types interact to produce rhythmic output. Modulators can then alter the relative contributions of the different pacemakers, leading to substantial changes in motor output and hence to different behaviors. Surprisingly, substantial changes in some aspects of the circuitry of a central pattern generator, such as a several-fold increase in synaptic strength, can sometimes have little effect on the output of the CPG, whereas other changes have profound effects.     

An artificial neural network that utilizes hip joint actuations to control bifurcations and chaos in a passive dynamic bipedal walking model

Chaos is a central feature of human locomotion and has been suggested to be a window to the control mechanisms of locomotion. In this investigation, we explored how the principles of chaos can be used to control locomotion with a passive dynamic bipedal walking model that has a chaotic gait pattern. Our control scheme was based on the scientific evidence that slight perturbations to the unstable manifolds of points in a chaotic system will promote the transition to new stable behaviors embedded in the rich chaotic attractor. Here we demonstrate that hip joint actuations during the swing phase can provide such perturbations for the control of bifurcations and chaos in a locomotive pattern. Our simulations indicated that systematic alterations of the hip joint actuations resulted in rapid transitions to any stable locomotive pattern available in the chaotic locomotive attractor. Based on these insights, we further explored the benefits of having a chaotic gait with a biologically inspired artificial neural network (ANN) that employed this chaotic control scheme. Remarkably, the ANN was quite robust and capable of selecting a hip joint actuation that rapidly transitioned the passive dynamic bipedal model to a stable gait embedded in the chaotic attractor. Additionally, the ANN was capable of using hip joint actuations to accommodate unstable environments and to overcome unforeseen perturbations. Our simulations provide insight on the advantage of having a chaotic locomotive system and provide evidence as to how chaos can be used as an advantageous control scheme for the nervous system.     

What does robotics offer animal behaviour?

There is a growing body of robot-based research that makes a serious claim to be a new methodology for biology. Robots can be used as models of specific animal systems to test hypotheses regarding the control of behaviour. At levels from learning algorithms to specific dendritic circuits, implementing a proposed controller in a robotic device tests it against real environments in a way that is difficult to simulate. This often provides insight into the true nature of the problem. It also enforces complete specifications and combines bodies of data. Current work can sometimes be criticized for drawing unjustified conclusions given the limited evaluation and inevitable inaccuracies of robot models. Nevertheless, this approach has led to novel hypotheses for animal behaviour and seems likely to provide fruitful results in the future. Copyright 2000 The Association for the Study of Animal Behaviour.     

A model of the cerebellar pathways applied to the control of a single-joint robot arm actuated by McKibben artificial muscles

 This article describes an expanded version of a previously proposed motor control scheme, based on rules for combining sensory and motor signals within the central nervous system. Classical control elements of the previous cybernetic circuit were replaced by artificial neural network modules having an architecture based on the connectivity of the cerebellar cortex, and whose functioning is regulated by reinforcement learning. The resulting model was then applied to the motion control of a mechanical, single-joint robot arm actuated by two McKibben artificial muscles. Various biologically plausible learning schemes were studied using both simulations and experiments. After learning, the model was able to accurately pilot the movements of the robot arm, both in velocity and position. Received: 4 September 2000 / Accepted in revised form: 7 November 2001     

CPG Control for Biped Hopping Robot in Unpredictable Environment

A CPG control mechanism is proposed for hopping motion control of biped robot in unpredictable environment.Based on analysis of robot motion and biological observation of animal's control mechanism,the motion control task is divided into two simple parts:motion sequence control and output force control.Inspired by a two-level CPG model,a two-level CPG control mechanism is constructed to coordinate the drivers of robot joint,while various feedback information are introduced into the control mechanism.Interneurons within the control mechanism are modeled to generate motion rhythm and pattern promptly for motion sequence control; motoneurons are modeled to control output forces of joint drivers in real time according to feedbacks.The control system can perceive changes caused by unknown perturbations and environment changes according to feedback information,and adapt to unpredictable environment by adjusting outputs of neurons.The control mechanism is applied to a biped hopping robot in unpredictable environment on simulation platform,and stable adaptive motions are obtained.     

Fly-like visuomotor responses of a robot using aVLSI motion-sensitive chips

 We explore the use of continuous-time analog very-large-scale-integrated (aVLSI) neuromorphic visual preprocessors together with a robotic platform in generating bio-inspired behaviors. Both the aVLSI motion sensors and the robot behaviors described in this work are inspired by the motion computation in the fly visual system and two different fly behaviors. In most robotic systems, the visual information comes from serially scanned imagers. This restricts the form of computation of the visual image and slows down the input rate to the controller system of the robot, hence increasing the reaction time of the robot. These aVLSI neuromorphic sensors reduce the computational load and power consumption of the robot, thus making it possible to explore continuous-time visuomotor control systems that react in real-time to the environment. The motion sensor provides two outputs: one for the preferred direction and the other for the null direction. These motion outputs are created from the aggregation of six elementary motion detectors that implement a variant of Reichardt's correlation algorithm. The four analog continuous-time outputs from the motion chips go to the control system on the robot which generates a mixture of two behaviors – course stabilization and fixation – from the outputs of these sensors. Since there are only four outputs, the amount of information transmitted to the controller is reduced (as compared to using a CCD sensor), and the reaction time of the robot is greatly decreased. In this work, the robot samples the motion sensors every 3.3 ms during the behavioral experiments. Received: 4 October 1999 / Accepted in revised form: 26 April 2001