Fish have a remarkable ability to maneuver in tight places, perform stable high acceleration maneuvers, hover efficiently, and quickly brake as a result of a complex muscular system that comprises more than half of the body mass. Additionally, fish have an extraordinary ability to sense minuscule changes in fluid flow through neuromasts in the lateral line which has been shown to allow fish to detect, localize, and track prey, perform synchronized schooling maneuvers, provide feedback control for efficient locomotion, and form hydrodynamic images of the environment which enable the fish to characterize entities in the vicinity. However, there is still very little understanding of the structure and organization of the hierarchical control systems or of how these actuation and sensing systems are integrated to perform steady and maneuvering locomotor tasks. Furthermore, there has been little effort to transform the biological concepts related to the sensing, actuation, and control of fish into truly bioinspired and biomimetic engineered materials and systems. This research aims to identify and theoretically describe the computational processing performed at the local sensory level for muscle activation and vertebral-stiffness modulation along the tail structure of fish for locomotion. Through a series of interdisplinary engineered experiments, the research seeks to understand (a) the ability of fish to actively modulate the mechanical properties of the tail via muscle recruitment, (b) how swimming gaits are regulated by a hierarchy of control systems that involve the visual, vestibular, and neuromast sensory systems, and (c) how hydrodynamic stimuli to the lateral line neuromasts directly influence the mechanical properties of the tail. An advanced multifunctional material system having distributed actuation and sensing will be developed to serve as a platform for validation and to provide greater understanding of the biology of these systems. The new material system will utilize innovative artificial neuromasts (sensors) and muscles (actuators) that are distributed and arranged as inspired by the configuration found in fish. The artificial neuromast will consist of a cluster of nanowires acting as hairs attached to ionic polymer artificial neurons to create robust, flexible, sensitive, and dynamically responsive sensors for fluid flow detection. The biologically inspired actuation provided by the multifunctional material utilizes a distribution of micron flexible matrix composite actuators in the material system. Through coupling of the biological and engineering experiments of the fish and artificial material system, the interdisplinary team will work together to develop a new framework for observing, identifying, and predicting the sensorimotor behavior of fish for locomotion and stiffness modulation. This research will advance the state-of-the-art development of multifunctional materials, leading to new structures that can intelligently sense and actuate a network of distributed robust sensors and actuators. Pioneer efforts include developing an advanced material system using nanotechnology and advanced composite technology, fabricating hierarchically structured sensors, creating new tools for bio-engineering investigations, and instigating a paradigm shift in the understanding of the organization and structure of the hierarchical control fish use for sensing and maneuvering.
Through collaborative efforts, an intellectual framework for education in K-12 classrooms, undergraduate research, and recruitment of minorities will be developed and implemented. One goal of the proposed education plan is to achieve broad impact on students? learning through dissemination of knowledge through K-12 programs at Harvard?s Museum of Natural History. A traveling exhibit will be developed on robotic fish that showcases the biology of aquatic propulsion, new actuator and sensing technologies and how these can be integrated to design a robotic fish. Assessment will establish measurable learning objectives and provide data on learning and improvement of the educational modules.
In this multi-university collaborative research project (Virginia Tech, Harvard University, and Drexel University), we developed an engineering framework for the design and development of biologically inspired material systems and the synergistic effort of the team resulted in a greater understanding of the sensory and muscular systems of fish during locomotion. Understanding how fish move through the water and how they develop force and sense flows is a very promising avenue for developing next-generation underwater robotic devices that exceed the capabilities of current man-made devices such as propellers for underwater vehicles. Fish exhibit efficient and quiet locomotion and are able to accurately sense water flow around their bodies during swimming, capabilities that currently exceed human-engineered designs. Through the development of a robotic flapping system at Harvard (see figure), new findings into fish locomotion were discovered. For example, fish modulate their stiffness of their muscular system for different locomotor tasks and this device demonstrated that there is an optimal stiffness, length, and shape of the fin for different swimming speeds and maneuvers. Axial oscillations of the center of mass (COM) in fishes are not well modeled by conventional robotic devices which have substantially greater inertia when attached to laboratory measuring devices. Our new experiments showed that we could greatly reduce the within-cycle axial force oscillations with an appropriately phased imposed axial oscillation. Caudal and pectoral fin responses on fish using a vortex perturber developed as part of this research provided new clues into fin stiffness modulation during swimming. For example, it was found that fish increase the caudal fin stiffness with increases in swimming speed. Additionally, the hydrodynamic advantage of shark skin is not well understood and this robotic system using real shark skin and artificial shark skin (see figure) demonstrated that swimming shark skin has an improved performance for specified swimming motions and the shark skin showed a reduced static drag as compared to a smooth surface. Using digital image correlation (DIC) with artificial fins in a water tunnel, we found that there is an optimal joint and fin stiffness that maximizes swimming efficiency and thrust for different swimming conditions. These outcomes provide biologists and engineers with a greater understanding of the physiology of fish, and these outcomes can be beneficial for many government and commercial applications such as more efficient underwater propulsion systems and new skins and materials that reduce drag. Inspired by the neuromast found on fish, which is a hair bundle using for sensing, the team developed a new artificial hair cell sensory system using biomolecular materials (see figure). The artificial hair cell consists of a hair (cilium) embedded in a water-swollen hydrogel (follicle). Similar to living cells, the hair cell system uses a lipid bilayer for regulating ion transport and provides the electrical signal when the artificial hair is excited. Similarly new underwater thin-film flow sensors using carbonaceous nanomaterials have been developed. Using carbon nanohorns, nanotubes, and Gd3N@C80 filled nanotubes, the sensors provide an output voltage response that is proportional to the flow speed as well as directionality. The biologically inspired distributed sensing can be valuable for many applications, such as flow measurement for air and underwater vehicles, artificial hairs for advanced prosthetics and may lead to a new class of autonomous vehicles that can sense surrounding changes in water for hydrodynamic tracking and aid in propulsion optimization. This research had a significant impact on a number of students, research assistants, and postdoctoral fellows. The research provided the participants with interdisciplinary training and a unique research experience, and they learned the need and synergistic importance for bringing different disciplines together in understanding many engineering and physiological systems. Out of the 44 participants that were involved in the collaboration, 23 were underrepresented students and postdoctoral fellows. 15 of these students participated in research at each of the collaborating laboratories through the EFRI REM program in 2012 and 2013. These students presented their research during weekly meetings, the NSF grantees conferences, Virginia Tech Undergraduate Research Conference, and at professional conferences (NSF ERN and SPIE Smart Structures). As part of the EFRI grant, the renovation of the more than 50 year-old fishes gallery in the Harvard Museum of Natural History (HMNH) was completed. The HMNH now receives over 200,000 visitors each year with about 35,000 school children attending in groups to view the exhibits. The new exhibit contains a wealth of up-to-date information on fish biology, and includes a section on robotic fish with models representative of the range of activities undertaken on the EFRI grant. The new exhibit includes an interactive video kiosk with videos describing a range of research on fish biology and biomechanics and features EFRI research done in the Lauder lab.
