Triggered by the evolution of communication technology and the revitalization of manufacturing, society is becoming a society in which people do not exist in real space, but can engage in activities through avatars, their alter egos. In other words, it is necessary to consider the question of what we should do from the perspective of ``health'' in response to the previously unanticipated issue of communication through non-human entities. In response to the issue of how to maintain health and realize a happy and smiling society in the midst of activities in a different form than before, we have clarified what we should pay attention to in maintaining health, what we should think about and refer to in order to create a method of evaluating health, and what we should do in the past to improve our health. In addition, we would like to examine the possibility of using physiological indicators, as represented by wearable devices, and new health indicators using the latest analytical technologies.

When applying robotic hands to tasks in human environments, the hands must adapt to various shapes of objects to work with. This paper proposes a wire driving system that can easily control grasping tasks, by using electromagnetic clutches. Grasping control is carried out using the slippage of the clutches. In the proposed hand mechanism, a single motor can drive multiple fingers by using clutches that selectively connect the wire of each finger. The prototype hand is capable of object grasping with five-fingers using a single motor. As the grasping force is determined by friction force of the clutch, force sensors are not required. Furthermore, no overload occurs on the fingers or the grasped objects. In this paper, we first demonstrated grasping force control experiments by manipulating the clutch voltage. Then, grasping tests with five fingers are shown for objects of various shapes. Finally, we demonstrated through experiments that gesture control is also possible. Independent finger control is carried out by providing ON/OFF signals to each clutch accordingly to the desired bending angle.

Recently, jig-less robotic assembly has been desired in the manufacturing industry. In jig-less conditions, each part is assumed to be supplied anywhere in a compartment of a parts tray. In particular, if a part is close to the partitions or the corners of the tray, grasping with a versatile robotic hand is difficult. In this paper, we propose a robust grasping strategy for a cylindrical part at any position in a parts tray compartment to achieve a desired grasp using a versatile hand with parallel stick fingers, and confirm the feasibility of grasping by the proposed strategy through real experiments.

The construction of lunar infrastructure is crucial for future human activities. Unmanned vehicles, such as rovers, serve as key mobility systems for construction tasks but face the risk of becoming trapped by sand and immobilized on uneven terrains. This paper focuses on a scenario where a rover being stuck is rescued by another rover via a robotic manipulator. We design and prototype a two-link robotic manipulator and conduct dynamic simulation along with experimental demonstrations. The experimental result are generally consistent with simulation results, particularly in terms of identifying the energy-efficient configurations of the manipulator for effective rescue operations.

When robots are deployed in large environments such as hospitals and offices, they must learn place-object relationships in a short period of time. However, the amount of observational data required for multiple robots to perform object search and tidy-up tasks satisfactorily is often unclear a priori, making rapid knowledge acquisition necessary. Therefore, we propose a method in which each robot inputs its knowledge based on on-site learning of a spatial concept model into a large language model, GPT-4, to infer probabilistic action planning based on its predictions. We conducted simulations of object search tasks with multiple robots according to user instructions and evaluated the success score of each task for each iteration of spatial concept learning. As a result of the experiment, the proposed method achieved a high success score while reducing the amount of observational data by more than half compared to the baseline.

Regrasping is a crucial upstream process in many manipulation tasks. In particular, dynamic regrasping, which involves transitioning the grasping state by throwing an object, enables rapid operations but requires precise execution of throwing and catching. This study focuses on the throwing motion and proposes a two-stage strategy using a two-finger gripper and a robot arm based on an optimization method that considers joint constraints and successful catching. Simulation experiments, conducted under actual robot constraints, demonstrated that the strategy successfully achieved dynamic regrasping in 22 out of 23 target orientations.

As a simple first step in designing a robotic hand, the flection and extension of finger joints have been focused on building to mimic hand shape. However, there are many limitations to using these robotic hands to grasp and handle objects. In contrast, our hands have a more complex mechanism with two degrees of freedom at the carpometacarpal joints, allowing us to grasp the arch shape by deforming the palm. In this study, we presented one of the mechanical design ideas for realizing flexible functions in the metacarpal bones of the palm, and described the evaluation results using a prototype.

Cable manipulation is a challenging task for robots due to the complexity of its operating characteristics. In situations where the work environment changes, such as on a factory production line, unpredictable constraints on the cable can hinder operation and make it even more difficult. In this paper, we propose a method for detecting the constrained position of cables by integrating visual and force information. We focused on that a restrained cable has a restricted motion range and generates tension against the restrained position, and detected the restrained position with high accuracy based on the observation of cable motion and the information of tension. This proposal enables detection of the cable's constrained position even in an unlearned environment.