Robot Grasp Planning: A Learning from Demonstration-Based Approach.

Robot grasping constitutes an essential capability in fulfilling the complexities of advanced industrial operations. This field has been extensively investigated to address a range of practical applications. However, the generation of a stable grasp remains challenging, principally due to the constraints imposed by object geometries and the diverse objectives of the tasks. In this work, we propose a novel learning from demonstration-based grasp-planning framework. This framework is designed to extract crucial human grasp skills, namely the contact region and approach direction, from a single demonstration. Then, it formulates an optimization problem that integrates the extracted skills to generate a stable grasp. Distinct from conventional methods that rely on learning implicit synergies through human demonstration or on mapping the dissimilar kinematics between human hands and robot grippers, our approach focuses on learning the intuitive human intent that involves the potential contact regions and the grasping approach direction. Furthermore, our optimization formulation is capable of identifying the optimal grasp by minimizing the surface fitting error between the demonstrated contact regions on the object and the gripper finger surface and imposing a penalty for any misalignment between the demonstrated and the gripper's approach directions. A series of experiments is conducted to verify the effectiveness of the proposed algorithm through both simulations and real-world scenarios.

Robot Programming from a Single Demonstration for High Precision Industrial Insertion.

We propose a novel approach for robotic industrial insertion tasks using the Programming by Demonstration technique. Our method allows robots to learn a high-precision task by observing human demonstration once, without requiring any prior knowledge of the object. We introduce an Imitated-to-Finetuned approach that generates imitated approach trajectories by cloning the human hand's movements and then fine-tunes the goal position with a visual servoing approach. To identify features on the object used in visual servoing, we model object tracking as the moving object detection problem, separating each demonstration video frame into the moving foreground that includes the object and demonstrator's hand and the static background. Then a hand keypoints estimation function is used to remove the redundant features on the hand. The experiment shows that the proposed method can make robots learn precision industrial insertion tasks from a single human demonstration.

An optimized FACS-free single-nucleus RNA sequencing (snRNA-seq) method for plant science research.

Recently, single-cell RNA sequencing (scRNA-seq) provides unprecedented power for accurately understanding gene expression regulatory mechanisms. However, scRNA-seq studies have limitations in plants, due to difficulty in protoplast isolation that requires enzymatic digestion of the cell walls from various plant tissues. Therefore, to overcome this problem, we developed a nuclei isolation approach that does not rely on Fluorescence Activated Cell Sorting (FACS). We validated the robustness of the FACS-free single-nucleus RNA sequencing (snRNA-seq) methodology in mature Arabidopsis plant tissue by comparing it to scRNA-seq results based on protoplasts extracted from the same batch of leaf materials. Sequencing results demonstrated the high quality of snRNA-seq data, as well as its utility in cell type classification and marker gene identification. This approach also showed several advantages, including the ability to use frozen samples, taking less suspension preparation time, and reducing biased cellular coverage and dissociation-induced transcriptional artifacts. Surprisingly, snRNA-seq detected two epidermal pavement cell clusters, while scRNA-seq only had one. Furthermore, we hypothesized that these two epidermal cells represent the top and lower epidermis based on differences in expression patterns of cluster-specific expressed genes. In summary, this study has advanced the application of snRNA-seq in Arabidopsis leaves and confirmed the advantages of snRNA-seq in plant research.

Genome-Wide Analysis of the MADS-Box Gene Family in Holoparasitic Plants (Balanophora subcupularis and Balanophora fungosa var. globosa).

MADS-box is an important transcription factor family that is involved in the regulation of various stages of plant growth and development, especially flowering regulation and flower development. Being a holoparasitic plant, the body structure of Balanophoraceae has changed dramatically over time, and its vegetative and reproductive organs have been extensively modified, with rudimentary flower organs. Meanwhile, extraordinary gene losses have been identified in holoparasitic plants compared with autotrophs. Our study reveals that the MADS-box gene family contracted sharply in Balanophora subcupularis and Balanophora fungosa var. globosa, and some subfamilies were lost, exhibiting reduced redundancy in both. The genes that functioned in the transition from the vegetative to floral production stages suffered a significant loss, but the ABCE model genes remained intact. We further investigated genes related to flowering regulation in B. subcupularis and B. fungosa var. globosa, vernalization and autonomous ways of regulating flowering time remained comparatively integrated, while genes in photoperiod and circadian clock pathways were almost lost. Convergent gene loss in flowering regulation occurred in Balanophora and another holoparasitic plant Sapria himalayana (Rafflesiaceae). The genome-wide analysis of the MADS-box gene family in Balanophora species provides valuable information for understanding the classification, gene loss pattern, and flowering regulation mechanism of MADS-box gene family in parasitic plants.

Integrated All-Fiber Electronic Skin toward Self-Powered Sensing Sports Systems.

The development of wearable electronic skins (E-skins) requires devices with high flexibility, breathability, and antibacterial activity, as in sports sensing technology. Here, we report a flexible, breathable, and antibacterial triboelectric nanogenerator (TENG)-based E-skin for self-powered sensing in volleyball reception statistics and analytics, which is fabricated by sandwiching a silver nanowire (Ag NW) electrode between a thermoplastic polyurethane (TPU) sensing layer and a poly(vinyl alcohol)/chitosan (PVA/CS) substrate. Benefiting from an outstanding breathability of 10.32 kg m-2 day-1 and biocidal properties of CS and Ag NW, the E-skin offers excellent thermal-moisture comfort and a remarkable antibacterial effect on Escherichia coli and Staphylococcus aureus. A pressure sensitivity of 0.3086 V kPa-1 is demonstrated in the sensing range of 6.65-19.21 kPa. Besides, a volleyball reception statistical and analytical system is further developed based on a 2 × 3 E-skin array. According to this work, the integration of wearable electronic devices with self-powered sensors may expand practical applications in sports.

Molecular cloning and tissue distribution of fatty acid binding protein-3 in goldfish (Carassius auratus) and its mRNA expression in response to cadmium and PAMPs.

Fatty acid binding proteins (FABPs) are members of the conserved, multigene family of intracellular lipid binding proteins. In this study, the full-length cDNA of goldfish (Carassius auratus) FABP-3 (gfFABP-3) was successfully cloned. gfFABP-3 had an open reading frame of 402 bp and encoded a 133 amino acid polypeptide. The predicted gfFABP-3 protein included a lipocalin domain and displayed typical conserved FABP tertiary structures. Reverse transcription-PCR (RT-PCR) revealed that the gfFABP-3 gene was expressed in all tested tissues, with higher levels of expression in the testis, liver, heart, fat and kidney. After 24 h of cadmium exposure, gfFABP-3 was significantly upregulated in the gill, liver and spleen, but downregulated in the intestine, as compared to unexposed controls. gfFABP-3 expression was significantly downregulated in the spleen in goldfish challenged with LPS and Poly I:C. Our study provides a molecular characterization of goldfish FABP-3 and indicated that gfFABP-3 was potentially associated with the toxic effects of cadmium on lipid metabolism, and with the immune response to pathogenic infection.

Combination of intraoperative 5-aminolevulinic acid-induced fluorescence and 3-D MR imaging for guidance of robotic laser ablation for precision neurosurgery.

A combination of 5-aminolevulinic acid (5-ALA)-induced fluorescence and three-dimensional (3-D) magnetic resonance imaging (MRI) of a brain tumor has been incorporated into a robotic laser ablation neurosurgery system. 5-ALA is a non-fluorescent prodrug that leads to intracellular accumulation of fluorescent protoporphyrins IX (PpIX) in malignant glioma. The PpIX tends to accumulate in pathological lesions, and emits red fluorescence when excited by blue light. This fluorescence is illuminated with laser excitation, enables intra-operative identification of the position of a tumor and provides guidance for resection with laser photocoagulation. The information provided by the MRI is enhanced by the 5-ALA fluorescence data, and this enhanced information is integrated into a robotic laser ablation system. The accuracy of the fluorescent measurement of the tumor is improved using high-precision spectral analysis. The fluorescence assists in the detection of malignant brain tumors intraoperatively and improves their removal rate.