Abstract

Abstract

Abstract

Abstract

Abstract
Robots are increasingly present in our lives, sharing the workspace and tasks with human co-workers. However, existing interfaces for human-robot interaction / cooperation (HRI/C) have limited levels of intuitiveness to use and safety is a major concern when humans and robots share the same workspace. Many times, this is due to the lack of a reliable estimation of the human pose in space which is the primary input to calculate the human-robot minimum distance (required for safety and collision avoidance) and HRI/C featuring machine learning algorithms classifying human behaviours / gestures. Each sensor type has its own characteristics resulting in problems such as occlusions (vision) and drift (inertial) when used in an isolated fashion. In this paper, it is proposed a combined system that merges the human tracking provided by a 3D vision sensor with the pose estimation provided by a set of inertial measurement units (IMUs) placed in human body limbs. The IMUs compensate the gaps in occluded areas to have tracking continuity. To mitigate the lingering effects of the IMU offset we propose a continuous online calculation of the offset value. Experimental tests were designed to simulate human motion in a human-robot collaborative environment where the robot moves away to avoid unexpected collisions with de human. Results indicate that our approach is able to capture the human's position, for example the forearm, with a precision in the millimetre range and robustness to occlusions.

Abstract
Several approaches with interesting results have been proposed over the years for robot grasp planning. However, the industry suffers from the lack of an intuitive and reliable system able to automatically estimate grasp poses while also allowing the integration of grasp information from the accumulated knowledge of the end user. In the presented paper it is proposed a non-object-agnostic grasping pipeline motivated by picking use cases from the aerospace industry. The planning system extends the functionality of the simulated annealing optimization algorithm for allowing its application within an industrial use case. Therefore, this paper addresses the first step of the design of a reconfigurable and modular grasping pipeline. The key idea is the creation of an intuitive and functional grasping framework for being used by factory floor operators according to the task demands. This software pipeline is capable of generating grasp solutions in an offline phase, and later on, in the robot operation phase, can choose the best grasp pose by taking into consideration a set of heuristics that try to achieve a successful grasp while also requiring the least effort for the robotic arm. The results are presented in a simulated and a real factory environment, relying on a mobile platform developed for intralogistic tasks. With this architecture, new state-of-art methodologies can be integrated in the future for growing the grasping pipeline and make it more robust and applicable to a wider range of use cases.