No abstract

No abstract

Wood vinegar, a by-product of wood pyrolysis, is mostly discarded. Wood vinegar has a phytotoxic effect and could be potentially used as a naturally derived herbicide for weed control. The objective of this research was to evaluate the efficacy of wood vinegar from the pyrolysis of apple (Malus × domestica Borkh.) tree branch wastes to control weeds. The wood vinegar concentrations required to inhibit 50% motherwort (Leonurus cardiaca L.), redroot pigweed (Amaranthus retroflexus L.), Spanish needles (Bidens pilosa L.), and tall fescue (Festuca arundinacea L.) seed germination measured 0.51%, 0.48%, 0.16%, and 1.1%, respectively. The wood vinegar application rates (spray volume) required to provide 50% control of motherwort and Spanish needles measured 1911 L ha−1 and 653 L ha−1, respectively, while the highest evaluated rate at 4000 L ha−1 controlled 35% tall fescue by 10 days after treatment (DAT). Common purslane (Portulaca oleracea L.) control increased as the wood vinegar application rate increased from 500 L ha−1 to 2000 L ha−1. Wood vinegar was more effective in dark than light conditions for controlling common purslane. By 5 DAT, averaged over application rates, wood vinegar provided 95% and 87% control of common purslane in dark and light conditions, respectively. These findings suggest that wood vinegar obtained from the pyrolysis of apple tree branches could be used for weed management.

No abstract

Despite the impressive progress achieved in robust grasp detection, robots are not skilled in sophisticated grasping tasks (e.g. search and grasp a specific object in clutter). Such tasks involve not only grasping, but comprehensive perception of the visual world (e.g. the relationship between objects). Recently, the advanced deep learning techniques provide a promising way for understanding the high-level visual concepts. It encourages robotic researchers to explore solutions for such hard and complicated fields. However, deep learning usually means data-hungry. The lack of data severely limits the performance of deep-learning-based algorithms. In this paper, we present a new dataset named REGRAD to sustain the modeling of relationships among objects and grasps. We collect the annotations of object poses, segmentations, grasps, and relationships in each image for comprehensive perception of grasping. Our dataset is collected in both forms of 2D images and 3D point clouds. Moreover, since all the data are generated automatically, users are free to import their own object models for the generation of as many data as they want. We have released our dataset and codes 1 . A video that demonstrates the process of data generation is also available 2 . * Equally contributed.