In the factory of the future, most of the operations will be done by autonomous robots that need visual feedback to move around the working space avoiding obstacles, to work collaboratively with humans, to identify and locate the working parts, to complete the information provided by other sensors to improve their positioning accuracy, etc. Different vision techniques, such as photogrammetry, stereo vision, structured light, time of flight and laser triangulation, among others, are widely used for inspection and quality control processes in the industry and now for robot guidance. Choosing which type of vision system to use is highly dependent on the parts that need to be located or measured. Thus, in this paper a comparative review of different machine vision techniques for robot guidance is presented. This work analyzes accuracy, range and weight of the sensors, safety, processing time and environmental influences. Researchers and developers can take it as a background information for their future works.

Automated surface defect detection is a challenging problem that has attracted major attention for decades. Traditional methods were designed using a pipeline of carefully designed operations. The resulting methods were complex systems, which were difficult to tune and adapt to different problems or data. A new approach to solving this problem has emerged recently: deep learning. This trend is motivated by two main factors: the increasing digitization of society, which makes it possible to record large datasets of labeled samples; and the availability of a large pool of computational resources. This work evaluates state-of-the-art deep-learning methods in object detection and semantic segmentation in the field of automated surface inspection in metals. Images acquired in the industry are affected by the conditions of the environment, including noise, dust, and vibrations, which is an additional challenge. Moreover, industrial inspection requires accuracy, but also robustness and speed. The selected methods are applied to different datasets of images that include the most common defects in metals and the performance is compared in terms of accuracy and speed. Results show exceptional accuracy at a fraction of the required processing time.

Intelligent automation, including robotics, is one of the current trends in the manufacturing industry in the context of “Industry 4.0”, where cyber-physical systems control the production at automated or semi-automated factories. Robots are perfect substitutes for a skilled workforce for some repeatable, general, and strategically-important tasks. However, this transformation is not always feasible and immediate, since certain technologies do not provide the required degree of flexibility. The introduction of collaborative robots in the industry permits the combination of the advantages of manual and automated production. In some processes, it is necessary to incorporate robots from different manufacturers, thus the design of these multi-robot systems is crucial to guarantee the maximum quality and efficiency. In this context, this paper presents a novel methodology for process automation design, enhanced implementation, and real-time monitoring in operation based on creating a digital twin of the manufacturing process with an immersive virtual reality interface to be used as a virtual testbed before the physical implementation. Moreover, it can be efficiently used for operator training, real-time monitoring, and feasibility studies of future optimizations. It has been validated in a use case which provides a solution for an assembly manufacturing process.

In this paper we present the utilization bound for Earliest Deadline First (EDF) scheduling on homogeneous multiprocessor systems with partitioning strategies. Assuming that tasks are pre-emptively scheduled on each processor according to the EDF algorithm, and allocated according to the First Fit (FF) heuristic, we prove that the worst-case achievable utilization is 0:5(n + 1), where n is the number of processors. This bound is valid for arbitrary utilization factors. Moreover, if all the tasks have utilization factors under a value α, the previous bound is raised, and the new utilization bound considering α is calculated. In addition, we prove that no pair of uniprocessor scheduling algorithm-allocation algorithm can provide a higher worst-case achievable utilization than that of EDF-FF. Finally, simulation provides the average-case achievable utilization for EDF-FF.