Search

282 p., Objektuak hartu eta uztea oinarrizko bi eragiketa dira ia edozein aplikazio robotikotan. Gaur egun, "pick and place" aplikazioetarako erabiltzen diren robot industrialek zeregin sinpleak eta errepikakorrak egiteko duten eraginkortasuna dute ezaugarri. Hala ere, sistema horiek oso zurrunak dira, erabat kontrolatutako inguruneetan lan egiten dute, eta oso kostu handia dakarte beste zeregin batzuk egiteko birprogramatzeak. Gaur egun, industria-ingurune desberdinetako zereginak daude (adibidez, logistika-ingurune batean eskaerak prestatzea), zeinak objektuak malgutasunez manipulatzea eskatzen duten, eta oraindik ezin izan dira automatizatu beren izaera dela-eta. Automatizazioa zailtzen duten botila-lepo nagusiak manipulatu beharreko objektuen aniztasuna, roboten trebetasun falta eta kontrolatu gabeko ingurune dinamikoen ziurgabetasuna dira.Adimen artifizialak (AA) gero eta paper garrantzitsuagoa betetzen du robotikaren barruan, robotei zeregin konplexuak betetzeko beharrezko adimena ematen baitie. Gainera, AAk benetako esperientzia erabiliz portaera konplexuak ikasteko aukera ematen du, programazioaren kostua nabarmen murriztuz. Objektuak manipulatzeko egungo sistema robotikoen mugak ikusita, lan honen helburu nagusia manipulazio-sistemen malgutasuna handitzea da AAn oinarritutako algoritmoak erabiliz, birprogramatu beharrik gabe ingurune dinamikoetara egokitzeko beharrezko gaitasunak emanez.

This paper describes the dynamic mosaic planning method developed in the context of the PICKPLACE European project. The dynamic planner has allowed the development of a robotic system capable of packing a wide variety of objects without having to adjust to each reference. The mosaic planning system consists of three modules: First, the picked item monitoring module monitors the grabbed item to find out how the robot has picked it. At the same time, the destination container is monitored online to obtain the actual status of the packaging. To this end, we present a novel heuristic algorithm that, based on the point cloud of the scene, estimates the empty volume inside the container as empty maximal spaces (EMS). Finally, we present the development of the dynamic IK-PAL mosaic planner that allows us to dynamically estimate the optimal packing pose considering both the status of the picked part and the estimated EMSs. The developed method has been successfully integrated in a real robotic picking and packing system and validated with 7 tests of increasing complexity. In these tests, we demonstrate the flexibility of the presented system in handling a wide range of objects in a real dynamic packaging environment. To our knowledge, this is the first time that a complete online picking and packing system is deployed in a real robotic scenario allowing to create mosaics with arbitrary objects and to consider the dynamics of a real robotic packing system.

La organización de diferentes objetos en una caja, conocido como Bin Packing, es un problema importante cuando se trata de automatizar la producción de un almacén. Hoy en día, a menudo se usan algoritmos que controlan un brazo robótico diciéndole donde poner cada objeto. Sin embargo, estas soluciones, debido a la complejidad del problema, rara vez pueden ser aplicadas en un caso práctico y por eso se ha querido mejorar una aplicación ya existente. En este trabajo se propone una nueva implementación basada en el control de los espacios libres que generan los objetos con el fin de reducir el coste computacional sin perder la calidad de los resultados. Los experimentos realizados muestran que este nuevo enfoque más eficiente que la aplicación original y que otros métodos alternativos.

Grasping point detection has traditionally been a core robotic and computer vision problem. In recent years, deep learning based methods have been widely used to predict grasping points, and have shown strong generalization capabilities under uncertainty. Particularly, approaches that aim at predicting object affordances without relying on the object identity, have obtained promising results in random bin-picking applications. However, most of them rely on RGB/RGB-D images, and it is not clear up to what extent 3D spatial information is used. Graph Convolutional Networks (GCNs) have been successfully used for object classification and scene segmentation in point clouds, and also to predict grasping points in simple laboratory experimentation. In the present proposal, we adapted the Deep Graph Convolutional Network model with the intuition that learning from n-dimensional point clouds would lead to a performance boost to predict object affordances. To the best of our knowledge, this is the first time that GCNs are applied to predict affordances for suction and gripper end effectors in an industrial bin-picking environment. Additionally, we designed a bin-picking oriented data preprocessing pipeline which contributes to ease the learning process and to create a flexible solution for any bin-picking application. To train our models, we created a highly accurate RGB-D/3D dataset which is openly available on demand. Finally, we benchmarked our method against a 2D Fully Convolutional Network based method, improving the top-1 precision score by 1.8% and 1.7% for suction and gripper respectively.

Teknibot robotarekin urrutiko elkarrekintzaren lehenengo pausuak diren Android aplikazioa. Aplikaziotik Tekniboti gida lanak egiteko aginduak bidaltzeko aukera, honen lokalizazioa ikusteko aukera eta robotak duen kameraren zenbait irudi ikusteko aukera ematen da.