Abstract

Abstract
While bottom-up approaches to object recognition are simple to design and implement, they do not yield the same performance as top-down approaches. On the other hand, it is not trivial to obtain a moderate number of plausible hypotheses to be efficiently verified by top-down approaches. To address these shortcomings, we propose a hybrid top-down bottom-up approach to object recognition where a bottom-up procedure that generates a set of hypothesis based on data is combined with a top-down process for evaluating those hypotheses. We use the recognition of rectangular cuboid shaped objects from 3D point cloud data as a benchmark problem for our research. Results obtained using this approach demonstrate promising recognition performances.

Abstract
This letter presents a novel architecture for evaluating the success of picking operations that are executed by industrial robots. It is formed by a cascade of machine learning algorithms (kNN and SVM) and uses information obtained by a 6 axis force/torque sensor and, if available, information from the built-in sensors of the robotic gripper. Beyond measuring the success or failure of the entire operation, this architecture makes it possible to detect in real-time when an object is slipping during the picking. Therefore, force and torque signatures are collected during the picking movement of the robot, which is decomposed into five different stages that allows to characterize distinct levels of success over time. Several trials were performed using an industrial robot with two different grippers for picking a long and flexible object. The experiments demonstrate the reliability of the proposed approach under different picking scenarios since, it obtained a testing performance (in terms of accuracy) up to 99.5% of successful identification of the result of the picking operations, considering an universe of 400 attempts.

Abstract
In the past few years, cable-driven robots have received some attention by the scientific community and the industry. They have special characteristics that made them very reliable to operate with the level of safeness that is required by different environments, such as, handling of hazardous materials in construction sites. This paper presents a cable-driven robot called SPIDERobot, that was developed for automated construction of architectural projects. This robot has a rotating claw and it is controlled by a set of 4 cables that allow 4 degrees of freedom. In addition to the robot, this paper introduces a Dynamic Control System (DCS) that controls the positioning of the robot and assures that the length of cables is always within a safe value. Results show that traditional force-feasible approaches are more influenced by the pulling forces or the geometric arrangement of all cables and their positioning is significantly less accurate than the DCS. Therefore, the architecture of the SPIDERobot is designed to enable an easily scaling up of the solution to higher dimensions for operating in realistic environments.

Abstract

Abstract
Nowadays, and considering flexibility, industrial robots still present some drawback that prevent them to be used in vast fields of the industry. One of their major limitations is related with their perception skills. In this area, and although the many developments verified on 3D object recognition systems in the research sphere, the number of solutions appearing in the industry level has been slow. Hence, this article tries to clarify some of the motives that difficult the technology transference (in what concerns object recognition) between both worlds. At the same time, it will be presented an industrial case scenario (inserted in an European Project) where some of the problems enumerated during the article are present.