Kazutoshi Tanaka

Robotic manipulators capable of regulating both compliance and stiffness offer enhanced operational safety and versatility. Here, we introduce Worm Gear-based Adaptive Variable Elasticity (WAVE), a variable stiffness actuator (VSA) that integrates a non-backdrivable worm gear. By decoupling the driving motor from external forces using this gear, WAVE enables precise force transmission to the joint, while absorbing positional discrepancies through compliance. WAVE is protected from excessive loads by converting impact forces into elastic energy stored in a spring. In addition, the actuator achieves continuous joint stiffness modulation by changing the spring's precompression length. We demonstrate these capabilities, experimentally validate the proposed stiffness model, show that motor loads approach zero at rest--even under external loading--and present applications using a manipulator with WAVE. This outcome showcases the successful decoupling of external forces. The protective attributes of this actuator allow for extended operation in contact-intensive tasks, and for robust robotic applications in challenging environments.

Cable-driven serpentine manipulators hold great potential in unstructured environments, offering obstacle avoidance, multi-directional force application, and a lightweight design. By placing all motors and sensors at the base and employing plastic links, we can further reduce the arm's weight. To demonstrate this concept, we developed a 9-degree-of-freedom cable-driven serpentine manipulator with an arm length of 545 mm and a total mass of only 308 g. However, this design introduces flexibility-induced variations, such as cable slack, elongation, and link deformation. These variations result in discrepancies between analytical predictions and actual link positions, making pose estimation more challenging. To address this challenge, we propose a physical reservoir computing based pose estimation method that exploits the manipulator's intrinsic nonlinear dynamics as a high-dimensional reservoir. Experimental results show a mean pose error of 4.3 mm using our method, compared to 4.4 mm with a baseline long short-term memory network and 39.5 mm with an analytical approach. This work provides a new direction for control and perception strategies in lightweight cable-driven serpentine manipulators leveraging their intrinsic dynamics.

The replication of object stiffness is essential for enhancing haptic feedback in virtual environments. However, existing research has overlooked how stylus stiffness influences the perception of virtual object stiffness during tool-mediated interactions. To address this, we conducted a psychophysical experiment demonstrating that changing stylus stiffness combined with visual stimuli altered users' perception of virtual object stiffness. Based on these insights, we developed Transtiff, a stylus-shaped interface capable of on-demand stiffness control using a McKibben artificial muscle mechanism. Unlike previous approaches, our method manipulates the perceived stiffness of virtual objects via the stylus by controlling the stiffness of the stylus without altering the properties of the real object being touched, creating the illusion of a hard object feeing soft. Our user study confirmed that Transtiff effectively simulates a range of material properties, such as sponge, plastic, and tennis balls, providing haptic rendering that is closely aligned with the perceived material characteristics. By addressing the challenge of delivering realistic haptic feedback through tool-based interactions, Transtiff represents a significant advancement in the haptic interface design for VR applications.

Understanding decision-making processes is informative for strategic planning. Aiming to understand human risk-taking behavior in decision-making, we investigate the possibility of classifying whether a shot is offensive or not, targeting table tennis videos. We define the problem in a multi-task setting: detecting shots with frame-level precision while classifying shot offensiveness, and, as an optional task, predicting which player made the shot. We use commercial table tennis videos for target analysis and propose audio-visual self-supervised training, leveraging web videos with similar camera views. Our local contrastive loss encourages the model to learn frame-wise action locality collaboratively with traditional segment-wise contrastive loss, which we call global contrastive loss. Experimental results proved that the collaboration of two contrastive losses boosts the prediction performance.

Imitation learning (IL) is a promising method for programming dual-arm manipulation easily by imitating demonstrations from human experts. However, IL for dual-arm manipulation is still challenging because operating two robotic arms to collect demonstrations requires considerable effort. Therefore, we present a novel IL framework for dual-arm manipulation: learning dual-arm manipulation from demonstration translated from a human and robotic arm (LfDT). LfDT collects demonstrations of one human and one robotic arm. Thus, a human expert can easily and precisely adjust its arm movements according to the movement of the robotic arm. LfDT collects demonstrations of one human and one robotic arm, whereas IL methods typically demand demonstrations of two robotic arms. Therefore, LfDT employs a domain-translation network to convert the demonstrations of one human and one robotic arm into demonstrations of two robotic arms, which are then used to learn dual-arm manipulation via IL. The experiments demonstrate that LfDT successfully converts the demonstrations and learns the dual-arm manipulation in both simulation and real-world.

We propose Transtiff, a stick-shaped device that can display various stiffness for stick-based haptic interaction. The device has a stiffness-changing joint replicating an artificial muscle mechanism in the relay portion of the stick to change its stiffness. Transtiff can be applied to touch interaction of the screen, augmenting the haptic experience of operating with a stylus pen, which is usually felt uniform. As applications, users can experience the sensation of pen and brush writing on a single device. In addition, it is possible to change the stiffness of the device for each object on the screen to reproduce the tactile feel of that object.

Human behaviors, with whole-body coordination, involve large-scale sensorimotor interaction. Spontaneous bodily movements in the early developmental stage potentially lead toward acquisition of such coordinated behavior. These movements presumably contribute to the structuration of sensorimotor interaction, providing specific regularities in bidirectional information among muscle activities and proprioception. Whether and how spontaneous movements, despite being task-free, structure and organize sensorimotor interactions in the entire body during early development remain unknown. Herein, to address these issues, we gained insights into the structuration process of the sensorimotor interaction in neonates and 3-mo-old infants. By combining detailed motion capture and musculoskeletal simulation, sensorimotor information flows among muscle activities and proprioception throughout the body were obtained. Subsequently, we extracted spatial modules and temporal state in sensorimotor information flows. Our approach demonstrated that early spontaneous movements elicited body-dependent sensorimotor modules, revealing age-related changes in them, depending on the combination or direction. The sensorimotor interactions also displayed temporal non-random fluctuations analogous to those seen in spontaneous activities in the cerebral cortex and spinal cord. Furthermore, we found recurring state sequence patterns across multiple participants, characterized by a substantial increase in infants compared to the patterns in neonates. Therefore, early spontaneous movements induce the spatiotemporal structuration in sensorimotor interactions and subsequent developmental changes. These results implicated that early open-ended movements, emerging from a certain neural substrate, regulate the sensorimotor interactions through embodiment and contribute to subsequent coordinated behaviors. Our findings also provide a conceptual linkage between early spontaneous movements and spontaneous neuronal activity in terms of spatiotemporal characteristics.

Factory automation robot systems often depend on specially-made jigs that precisely position each part, which increases the system's cost and limits flexibility. We propose a method to determine the 3D pose of an object with high precision and confidence, using only parallel robotic grippers and no parts-specific jigs. Our method automatically generates a sequence of actions that ensures that the real-world position of the physical object matches the system's assumed pose to sub-mm precision. Furthermore, we propose the use of “extrinsic” actions, which use gravity, the environment and the gripper geometry to significantly reduce or even eliminate the uncertainty about an object's pose. We show in simulated and real-robot experiments that our method outperforms our previous work, at success rates over 95%.

This study aimed to anticipate fractures of fragile food during robotic food manipulation. Anticipating fractures allows a robot to manipulate ingredients without irreversible failure. Food fracture models investigated in food texture fields explain the properties of fragile objects well. However, they may not directly apply to robot manipulation due to the variance in physical properties even within the same ingredient. To this end, we developed a fracture-anticipation system with a tactile sensing module and a simple recurrent neural network. The key idea was to allow the robot to break ingredients during training-sample collection. The timing of fractures was identified via simple signal processing and used for supervision. We performed real robot experiments with three typical fragile foods: tofu, potato chips, and bananas. As the first step toward flexible fragile-object manipulation, we evaluated the proposed method for the fundamental task of object picking. The method successfully grasped the fragile foods without fractures in an online demonstration. In an offline evaluation, the method predicted the fractures with a recall of approximately 80% for all ingredients with 60 breaking trials. We believe that our method can be used to avoid breakage in other types of food manipulation, e.g., holding, pressing, and rolling.