Search

Soft grippers are garnering increasing attention for their adeptness in conforming to diverse objects, particularly delicate items, without warranting precise force control. This attribute proves especially beneficial in unstructured environments and dynamic tasks such as food handling. Human hands, owing to their elevated dexterity and precise motor control, exhibit the ability to delicately manipulate complex food items, such as small or fragile objects, by dynamically adjusting their grasping configurations. Furthermore, with their rich sensory receptors and hand-eye coordination that provide valuable information involving the texture and form factor, real-time adjustments to avoid damage or spill during food handling appear seamless. Despite numerous endeavors to replicate these capabilities through robotic solutions involving soft grippers, matching human performance remains a formidable engineering challenge. Robotic competitions serve as an invaluable platform for pushing the boundaries of manipulation capabilities, simultaneously offering insights into the adoption of these solutions across diverse domains, including food handling. Serving as a proxy for the future transition of robotic solutions from the laboratory to the market, these competitions simulate real-world challenges. Since 2021, our research group has actively participated in RoboSoft competitions, securing victories in the Manipulation track in 2022 and 2023. Our success was propelled by the utilization of a modified iteration of our Retractable Nails Soft Gripper (RNSG), tailored to meet the specific requirements of each task. The integration of sensors and collaborative manipulators further enhanced the gripper’s performance, facilitating the seamless execution of complex grasping tasks associated with food handling. This article encapsulates the experiential insights gained during the application of our highly versatile soft gripper in these competition environments.

In additive manufacturing of metals, numerous techniques have been employed to sense print defects. Among these, acoustic-based sensing has the advantage of low cost and shows the most potential to identify both external and internal defects as an in-situ monitoring system. Using acoustic signals, researchers have broadly investigated non-machine learning and machine learning-based approaches to identify defects like balling, micro defects, lack of fusion pores, keyhole pores, cracks, and porosity. While most of these works have shown promising results for laser-based AM systems, few have explored how acoustic signals can be used effectively for Wire Arc Additive Manufacturing (WAAM) defect detection. This paper proposes a methodology to construct machine learning (ML)-based models on identifying geometrically defective bead segments using acoustic signals during the WAAM process. Geometrically defective bead segment or geometric defect is a defect that causes voids in the final printed part due to incomplete fusion between two non-uniform overlapping bead segments. Such a defect is currently not explored in the literature. The proposed methodology uses a novel dataset labeling approach to identify good and bad bead segments based on an optimal threshold of the range of mean curvature. Furthermore, the methodology targets defective bead segments based on acoustic feature inputs like Principal Components (PC) or Mel Frequency Cepstral Coefficients (MFCC). To understand the resulting performance of the defect identification models constructed based on the proposed methodology, experiments are performed and tested on a variety of ML models (KNN, SVM, RF, NN, and CNN) based on the Inconel 718 material. The results show that the combinatorics of two acoustic input features and five ML models can be able to identify geometrically defective segments accurately with F1 score that ranges from 80% to 85%.

Herein, the development of a multimodal, nature‐inspired unmanned aerial vehicle (UAV) that operates in three different flight regimes (rotor wing, tailsitter, cruise), unlike typical UAVs which only consist of two, is discussed. The platform uses a nature‐inspired method to achieve efficient hover, yet possesses the flexibility to enter a more agile state via additional tailsitter and cruise modes. Both the mechanical configuration and software/control architecture used to achieve three‐mode capability are documented in detail. A sigmoid blending control is implemented as the transition control strategy, consisting of transition coordinators that adjust the weight of each individual controller on the actuator outputs. To improve the transition sequence based on performance goals such as reduced altitude variation and throttle usage, an optimization routine is performed to obtain the optimized blending parameters. The optimized parameters are then experimentally verified on a physical prototype in lab conditions and shown to exhibit at least a 20% improvement for altitude variation and at least 5% less throttle usage across 4 consecutive transitions compared with baseline. Three‐mode capability outdoors in real‐world conditions are also demonstrated in multiple flights. Power draw in rotor‐wing mode is found to be 30% less than in tailsitter mode.

Due to their efficient locomotion and natural tolerance to hazardous environments, spherical robots have wide applications in security surveillance, exploration of unknown territory and emergency response. Numerous studies have been conducted on the driving mechanism, motion planning and trajectory tracking methods of spherical robots, yet very limited studies have been conducted regarding the obstacle avoidance capability of spherical robots. Most of the existing spherical robots rely on the “hit and run” technique, which has been argued to be a reasonable strategy because spherical robots have an inherent ability to recover from collisions. Without protruding components, they will not become stuck and can simply roll back after running into bstacles. However, for small scale spherical robots that contain sensitive surveillance sensors and cannot afford to utilize heavy protective shells, the absence of obstacle avoidance solutions would leave the robot at the mercy of potentially dangerous obstacles. In this paper, a compact magnetic field-based obstacle detection and avoidance system has been developed for miniature spherical robots. It utilizes a passive magnetic field so that the system is both compact and power efficient. The proposed system can detect not only the presence, but also the approaching direction of a ferromagnetic obstacle, therefore, an intelligent avoidance behavior can be generated by adapting the trajectory tracking method with the detection information. Design optimization is conducted to enhance the obstacle detection performance and detailed avoidance strategies are devised. Experimental results are also presented for validation purposes.