Herein, we proposed an "animal-in-the-loop" (AIL) system by introducing a robot and a virtual reality (VR) technology in a conventional insect behavior experiment. The setup provided sensory inputs to an insect, which mimic its natural environment and simultaneously measured the behavioral output. The proposed AIL system consisted of a multimodal VR device and a ground-running robot, both of which were connected wirelessly. The insect behavior was measured using a multimodal VR device, and the behavioral changes were transmitted to the robot as control inputs. Specifically, the multimodal VR device was equipped with three types of sensory stimulators, odor, wind, and vision, and each stimulator was controlled by the value of the corresponding sensor on the robot. The surrounding environment was observed using multiple sensors mounted on the robot, and the information was transmitted to the VR device to provide sensory stimuli to the insect. This system allowed the insect on the VR device to remotely control the robot and perform localization virtually. The localization trajectories of the proposed AIL system were similar to those of the free-walking experiment, and the tendency of the change in the heading angle during localization was also similar. Therefore, we found that using the AIL system enabled us to measure behavioral changes upon providing sensory stimuli to insects. These VR stimuli were similar to those encountered by the insect in free-walking experiments.

The goal of our research is to give the robot environmental adaptability. As a result of pursuing accuracy, the conventional robot is rigid and sturdy, and can operate at high speed with high precision in a well-known space. In contrast, these systems have difficulty in unknown or unstructured environments. We thought that the "flexibility" of living things is the key to solve this problem for robots of the future. Living things process a variety of stimuli, act accordingly, and sometimes change their bodies to adapt to the environment. We defined the "flexibility" of a living being as its intelligence, movement, and body. Muscle cells are one of the candidate materials that enable "flexible" robots. It has been reported that myoblasts, the material of muscle cells, fuse by induction of differentiation, and their properties change depending on the growth environment. Therefore, muscle cells have not only physically flexible but also environmental adaptability. Furthermore, the advent of 3D printing technology in recent years has enabled us to freely create three-dimensional structures and has greatly contributed to the development of conventional robots. In fact, the development of an actuator composed of muscle cells (muscle-cell based actuator) with a 3D printer has been reported. On the other hand, only a part of the robot has been replaced with cells, and the production requires the knowledge and skills of some engineers, so it has not been put to practical use. Previous studies have reported that muscle cells can be used as pressure sensors. For this reason, muscle cells seem to become all the CPUs, sensors, and actuators that compose a robot. However, their hierarchical structure and performance are unclear for a muscle-cell robot. Thus, we aim to establish 4D printing technology that can embed dynamic elements (like muscle cells) into artificial objects. In this study, we defined 4D printing technology as printing technology that adds dynamic elements of cells to 3D printing. For the establishment of 4D printing technology, we examined the method of arranging muscle cells and selected gel working as a scaffold for muscle cells in this report. We tried two methods of cell arrangement. One is the attachment method and the other is the embedding method. The attachment method is first printing only the gel and then spreading the cells on the surface of the gel. The embedding method is to print the gel containing the cells. We also used two gels, GelMA (CELLINK) and Collagen 1A (Nitta Gelation) and evaluated adhesion (whether cells can stay in the gel), printability, differentiation, and shape retention. These experiments were performed under four conditions combined with the cell arrangement methods and the gel. As a result, cells could not be arranged evenly by the attachment method, whereas the embedding method did not have that problem. Collagen 1A was seem to be better for the gel, because it contributed to cell differentiation and shape maintenance, but it had slightly lower printing properties than GelMA. Finally, we examined whether the gel with printed myoblast functions as a muscle-cell based actuator. It was observed that the gel containing muscle cells contracted in response to electrical stimulation. These data suggested that printed myoblast by a 3D printer functions as an actuator. We concluded that our method can produce functional muscle cells. Towards establishing 4D printing technology, we will investigate the arrangement and function of the printed cells to be printed. We plan to clarify the optimal combination for creating sensors, actuators, and CPUs, which are the components of a muscle cell robot, by constructing a robot changing the combination of the structure and processing method of cells.

No abstract