Hillel J. Chiel

The brain has a body: adaptive behavior emerges from interactions of nervous system, body and environment

Studies of mechanisms of adaptive behavior generally focus on neurons and circuits. But adaptive behavior also depends on interactions among the nervous system, body and environment: sensory preprocessing and motor post-processing filter inputs to and outputs from the nervous system; co-evolution and co-development of nervous system and periphery create matching and complementarity between them; body structure creates constraints and opportunities for neural control; and continuous feedback between nervous system, body and environment are essential for normal behavior. This broader view of adaptive behavior has been a major underpinning of ecological psychology and has influenced behavior-based robotics. Computational neuroethology, which jointly models neural control and periphery of animals, is a promising methodology for understanding adaptive behavior.

In vivo buccal nerve activity that distinguishes ingestion from rejection can be used to predict behavioral transitions in Aplysia.

Abstract1.We are studying the neural basis of consummatory feeding behavior in Aplysia using intact, freely moving animals.2.Video records show that the timing of radula closure during the radula protraction-retraction cycle constitutes a major difference between ingestion (biting or swallowing) and rejection. During ingestion, the radula is closed as it retracts. During rejection, the radula is closed as it protracts.3.We observed two patterns of activity in nerves which are likely to mediate these radula movements. Patterns I and II are associated with ingestion and rejection, respectively, and are distinguished by the timing of radula nerve activity with respect to the onset of buccal nerve 2 activity.4.The association of ingestion with pattern I is maintained when the animal feeds on a polyethylene tube, the same food substrate used to elicit rejection responses. Under these conditions, pattern I is associated with either swallowing or no net tube movement.5.Most transitions from swallowing to rejection were preceded by one or more occurrences of pattern I in which there was no net tube movement, suggesting that these transitions can be predicted.6.Our data suggest that these two patterns can be used to distinguish ingestion from rejection.

Biologically based distributed control and local reflexes improve rough terrain locomotion in a hexapod robot

Distributed control and local leg reflexes enable insects to cope easily with terrain that would defeat many legged robots. An insect-like hexapod robot incorporating biologically based control effectively responded to mechanical perturbations using active and passive compliance and a local stepping reflex. An elevator reflex and a searching reflex addressed unexpected obstacles and loss of support, respectively. The robot exhibited a range of gaits using stick-insect-based distributed control mechanisms and negotiated irregular, slatted and compliant surfaces with this biologically based control strategy.

The timing of activity in motor neurons that produce radula movements distinguishes ingestion from rejection in Aplysia

Abstract1.We have studied the neural circuitry mediating ingestion and rejection in Aplysia using a reduced preparation that produces ingestion-like and rejection-like motor patterns in response to physiological stimuli.2.We have characterized 3 buccal ganglion motor neurons that produce specific movements of the radula and buccal mass. B8a and B8b act to close the radula. B10 acts to close the jaws and retract the radula.3.The patterns of activity in these neurons can be used to distinguish the ingestion-like and rejection-like motor patterns. B8a, B8b and B10 are active together during the ingestion-like pattern. Activity in B8a and B8b ends prior to the onset of activity in B10 during the rejection-like pattern.4.Our data suggest that these neurons undergo similar patterns of activity in vivo. During both feeding-like patterns, the activity and peripheral actions of B8a, B8b, and B10 are consistent with radula movements observed during ingestion and rejection. In addition, the extracellular activity produced by these neurons is consistent with neural activity observed in vivo during ingestion and rejection.5.Our data suggest that the different activity patterns observed in these motor neurons contribute to the different radula movements that distinguish ingestion from rejection.

Using autonomous robotics to teach science and engineering

f you walked into our Autonomous Robotics class at Case Western Reserve University on a typical day, you might be surprised to find 30 college students from a variety of engineering and science disciplines sitting on the floor surrounded by LEGO blocks. Appearances can be deceiving; this course tackles serious issues in engineering and science education. In this course, students design, build, program and test their own autonomous robots that participate in a public competition. This course uses robotics to foster a hands-on, interdisciplinary, teamwork-oriented approach to the synthesis and analysis of integrated real-world systems, as well as teaching new approaches to robot control. Created in 1995, our Autonomous Robotics course grew out of ongoing research on biologically inspired robotics at CWRU [3]. The design of this course draws heavily on technology developed at MIT, first for K–12 education [11] and later for an undergraduate course similar to ours that has been offered since 1990 [8]. Related courses have been developed at the University of Edinburgh [7] and elsewhere. Two features of our course distinguish it from these other courses. First, our final competition is considerably more technically demanding. Second, we address a much broader set of educational goals. Our course attracts students from computer engineering and science, biology, electrical engineering, neuroscience, systems engineering, biomedical engineering, and physics. In this article, we describe the educational goals of the course, its overall design, the final competition, and student assessment.

Evolution and Analysis of Model CPGs for Walking: II. General Principles and Individual Variability

Are there general principles for pattern generation? We examined this question by analyzing the operation of large populations of evolved model central pattern generators (CPGs) for walking. Three populations of model CPGs were evolved, containing three, four, or five neurons. We identified six general principles. First, locomotion performance increased with the number of interneurons. Second, the top 10 three-, four-, and five-neuron CPGs could be decomposed into dynamical modules, an abstract description developed in a companion article. Third, these dynamical modules were multistable: they could be switched between multiple stable output configurations. Fourth, the rhythmic pattern generated by a CPG could be understood as a closed chain of successive destabilizations of one dynamical module by another. A combinatorial analysis enumerated the possible dynamical modular structures. Fifth, one-dimensional modules were frequently observed and, in some cases, could be assigned specific functional roles. Finally, dynamic dynamical modules, in which the modular structure itself changed over one cycle, were frequently observed. The existence of these general principles despite significant variability in both patterns of connectivity and neural parameters was explained by degeneracy in the maps from neural parameters to neural dynamics to behavior to fitness. An analysis of the biomechanical properties of the model body was essential for relating neural activity to behavior. Our studies of evolved model circuits suggest that, in the absence of other constraints, there is no compelling reason to expect neural circuits to be functionally decomposable as the number of interneurons increase. Analyzing idealized model pattern generators may be an effective methodology for gaining insights into the operation of biological pattern generators.

A distributed neural network architecture for hexapod robot locomotion

We present fully distributed neural network architecture for controlling the locomotion of a hexapod robot. The design of this network is directly based on work on the neuroethology of insect locomotion. Previously, we demonstrated in simulation that this controller could generate a continuous range of statically stable insect-like gaits as the activity of a single command neuron was varied and that it was robust to a variety of lesions. We now report that the controller can be utilized to direct the locomotion of an actual six-legged robot, and that it exhibits a range of gaits and degree of robustness in the real world that is quite similar to that observed in simulation.

Development of a peristaltic endoscope

A device that could locomote through curving and tortuous spaces would find many applications in medicine and in industry. Invertebrates such as earthworms and leeches can solve this problem using peristaltic locomotion. We describe a device consisting of three braided pneumatic actuators in series that can successfully locomote peristaltically. The device can locomote forwards and backwards in elevated and curving tubes, and with a plastic sheath around it.

The Brain in Its Body: Motor Control and Sensing in a Biomechanical Context

Although it is widely recognized that adaptive behavior emerges from the ongoing interactions among the nervous system, the body, and the environment, it has only become possible in recent years to experimentally study and to simulate these interacting systems. We briefly review work on molluscan feeding, maintenance of postural control in cats and humans, simulations of locomotion in lamprey, insect, cat and salamander, and active vibrissal sensing in rats to illustrate the insights that can be derived from studies of neural control and sensing within a biomechanical context. These studies illustrate that control may be shared between the nervous system and the periphery, that neural activity organizes degrees of freedom into biomechanically meaningful subsets, that mechanics alone may play crucial roles in enforcing gait patterns, and that mechanics of sensors is crucial for their function.

Robustness of a distributed neural network controller for locomotion in a hexapod robot

The robustness of a distributed neural-network controller for locomotion based on insect neurobiology has been used to control a hexapod robot. The robustness of the controller is investigated experimentally. Disabling any single sensor, effector, or central component did not prevent the robot from walking. Furthermore, statically stable gaits could be established using either sensor input or central connections. Thus, a complex interplay between central neural elements and sensor inputs is responsible for the robustness of the controller and its ability to generate a continuous range of gaits. These results suggest that biologically inspired neural-network controllers may be a robust method for robotic control. >