Diverse Motion In-betweening from Sparse Keyframes with Dual Posture Stitching.

In-betweening is a technique for generating transitions given start and target character states. The majority of existing works require multiple (often ≥ 10) frames as input, which are not always available. In addition, they produce results that lack diversity, which may not fulfill artists' requirements. Addressing these gaps, our work deals with a focused yet challenging problem: generating diverse and high-quality transitions given exactly two frames (only the start and target frames). To cope with this challenging scenario, we propose a bi-directional motion generation and stitching scheme which generates forward and backward transitions from the start and target frames with two adversarial autoregressive networks, respectively, and stitches them midway between the start and target frames. In contrast to stitching at the start or target frames, where the ground truth cannot be altered, there is no strict midway ground truth. Thus, our method can capitalize on this flexibility and generate high-quality and diverse transitions simultaneously. Specifically, we employ conditional variational autoencoders (CVAEs) to implement our autoregressive networks and propose a novel stitching loss to stitch the bi-directional generated motions around the midway point.

Soft Actuator with Programmable Design: Modeling, Prototyping, and Applications.

Designs of soft actuators are mostly guided and limited to certain target functionalities. This article presents a novel programmable design for soft pneumatic bellows-shaped actuators with distinct motions, thus a wide range of functionalities can be engendered through tuning channel parameters. According to the design principle, a kinematic model is established for motion prediction, and a sampling-based optimal parameter search is executed for automatic design. The proposed design method and kinematic models provide a tool for the generation of an optimal channel curve, with respect to target functions and required motion trajectories. Quantitative characterizations on the analytical model are conducted. To validate the functionalities, we generate three types of actuators to cover a wide range of motions in manipulation and locomotion tasks. Comparisons of model prediction on motion trajectory and prototype performance indicate the efficacy of the forward kinematics, and two task-based optimal designs for manipulation scenarios validate the effectiveness of the design parameter search. Prototyped by additive manufacturing technique with soft matter, multifunctional robots in case studies have been demonstrated, suggesting adaptability of the structure and convenience of the soft actuator's automatic design in both manipulation and locomotion. Results show that the novel design method together with the kinematic model paves a way for designing function-oriented actuators in an automatic flow.

Multidimensional Tactile Sensor with a Thin Compound Eye-Inspired Imaging System.

Artificial tactile sensing for robots is a counterpart to the human sense of touch, serving as a feedback interface for sensing and interacting with the environment. A vision-based tactile sensor has emerged as a novel and advantageous branch of artificial tactile sensors. Compared with conventional tactile sensors, vision-based tactile sensors possess stronger potential thanks to acquiring multimodal contact information in much higher spatial resolution, although they typically suffer from bulky size and fabrication challenges. In this article, we report a thin vision-based tactile sensor that draws inspiration from natural compound eye structures and demonstrate its capability of sensing three-dimensional (3D) force. The sensor is composed of an array of vision units, an elastic touching interface, and a supporting structure with illumination. Experiments validated the sensor's advantages, including competitive spatial resolution of deformation as high as 1016 dpi on a 5 × 8 mm2 sensing area, superior accuracy of 3D force measurement at levels of 0.018 N for tangential force and 0.213 N (0.108 N at the center region) for normal force, and real-time processing at 30 Hz, while achieving a thin size of 5 mm. We further demonstrate the sensor capability in sensing 3D force and slip occurrence in real grasping experiments. This device paves the way for robotic applications that require rich tactile information with miniaturized sensor structure.

Hybrid Jamming for Bioinspired Soft Robotic Fingers.

This article describes a novel design of bioinspired soft robotic fingers based upon hybrid jamming principle-integrated layer jamming and particle jamming. The finger combines a fiber-reinforced soft pneumatic actuator with a hybrid jamming substrate. Taking advantage of different characteristics of layer jamming and particle jamming, the substrate is designed with three chambers filled with layers (function as bones) and two chambers filled with particles (function as joints). The layer regions and particle regions are interlocked with each other to guarantee load transfer from the fixed finger end to fingertip. With the proposed design, the finger is endowed with bending shape control, as well as variable stiffness capabilities. Theoretical analysis is conducted to predict the stiffness variation of the proposed finger at different vacuum levels, and experimental tests are performed to evaluate the finger's shape control and stiffness tuning effectiveness. Experimental results show that the proposed finger can achieve 5.52 times stiffness enhancement at primary position. Finally, we fabricate a gripper and perform grasping demonstrations on several objects. Results show that the gripper is able to transfer between low stiffness state for adaptive grasping and high stiffness state for robust holding.