Abstract

Abstract
Robotics will significantly impact large sectors of the economy with relatively low productivity, such as Agri-Food production. Deploying agricultural robots on the farm is still a challenging task. When it comes to localising the robot, there is a need for a preliminary map, which is obtained from a first robot visit to the farm. Mapping is a semi-autonomous task that requires a human operator to drive the robot throughout the environment using a control pad. Visual and geometric features are used by Simultaneous Localisation and Mapping (SLAM) Algorithms to model and recognise places, and track the robot's motion. In agricultural fields, this represents a time-consuming operation. This work proposes a novel solution-called AgRoBPP-bridge-to autonomously extract Occupancy Grid and Topological maps from satellites images. These preliminary maps are used by the robot in its first visit, reducing the need of human intervention and making the path planning algorithms more efficient. AgRoBPP-bridge consists of two stages: vineyards row detection and topological map extraction. For vineyards row detection, we explored two approaches, one that is based on conventional machine learning technique, by considering Support Vector Machine with Local Binary Pattern-based features, and another one found in deep learning techniques (ResNET and DenseNET). From the vineyards row detection, we extracted an occupation grid map and, by considering advanced image processing techniques and Voronoi diagrams concept, we obtained a topological map. Our results demonstrated an overall accuracy higher than 85% for detecting vineyards and free paths for robot navigation. The Support Vector Machine (SVM)-based approach demonstrated the best performance in terms of precision and computational resources consumption. AgRoBPP-bridge shows to be a relevant contribution to simplify the deployment of robots in agriculture.

Abstract
How to visualize recorded production data in Virtual Reality? How to use state of the art Augmented Reality displays that can show robot data? This paper introduces an open-source ICT framework approach for combining Unity-based Mixed Reality applications with robotic production equipment using ROS Industrial. This publication gives details on the implementation and demonstrates the use as a data analysis tool in the context of scientific exchange within the area of Mixed Reality enabled human-robot co-production.

Abstract
Purpose This paper aims to propose a translation library capable of generating robots proprietary code after their offline programming has been performed in a software application, named AdaptPack Studio, running over a robot simulation and offline programming software package. Design/methodology/approach The translation library, named AdaptPack Studio Translator, is capable to generate proprietary code for the Asea Brown Boveri, FANUC, Keller und Knappich Augsburg and Yaskawa Motoman robot brands, after their offline programming has been performed in the AdaptPack Studio application. Findings Simulation and real tests were performed showing an improvement in the creation, operation, modularity and flexibility of new robotic palletizing systems. In particular, it was verified that the time needed to perform these tasks significantly decreased. Practical implications The design and setup of robotics palletizing systems are facilitated by an intuitive offline programming system and by a simple export command to the real robot, independent of its brand. In this way, industrial solutions can be developed faster, in this way, making companies more competitive. Originality/value The effort to build a robotic palletizing system is reduced by an intuitive offline programming system (AdaptPack Studio) and the capability to export command to the real robot using the AdaptPack Studio Translator. As a result, companies have an increase in competitiveness with a fast design framework. Furthermore, and to the best of the author's knowledge, there is also no scientific publication formalizing and describing how to build the translators for industrial robot simulation and offline programming software packages, being this a pioneer publication in this area.

Abstract
Purpose This paper aims to propose an automated framework for agile development and simulation of robotic palletizing cells. An automatic offline programming tool, for a variety of robot brands, is also introduced. Design/methodology/approach This framework, named AdaptPack Studio, offers a custom-built library to assemble virtual models of palletizing cells, quick connect these models by drag and drop, and perform offline programming of robots and factory equipment in short steps. Findings Simulation and real tests performed showed an improvement in the design, development and operation of robotic palletizing systems. The AdaptPack Studio software was tested and evaluated in a pure simulation case and in a real-world scenario. Results have shown to be concise and accurate, with minor model displacement inaccuracies because of differences between the virtual and real models. Research limitations/implications An intuitive drag and drop layout modeling accelerates the design and setup of robotic palletizing cells and automatic offline generation of robot programs. Furthermore, A* based algorithms generate collision-free trajectories, discretized both in the robot joints space and in the Cartesian space. As a consequence, industrial solutions are available for production in record time, increasing the competitiveness of companies using this tool. Originality/value The AdaptPack Studio framework includes, on a single package, the possibility to program, simulate and generate the robot code for four different brands of robots. Furthermore, the application is tailored for palletizing applications and specifically includes the components (Building Blocks) of a particular company, which allows a very fast development of new solutions. Furthermore, with the inclusion of the Trajectory Planner, it is possible to automatically develop robot trajectories without collisions.

Abstract
Deep learning (DL) incorporates a modern technique for image processing and big data analysis with large potential. Deep learning is a recent tool in the agricultural domain, being already successfully applied to other domains. This article performs a survey of different deep learning techniques applied to various agricultural problems, such as disease detection/identification, fruit/plants classification and fruit counting among other domains. The paper analyses the specific employed models, the source of the data, the performance of each study, the employed hardware and the possibility of real-time application to study eventual integration with autonomous robotic platforms. The conclusions indicate that deep learning provides high accuracy results, surpassing, with occasional exceptions, alternative traditional image processing techniques in terms of accuracy.