Automatic, situation-based control of multirobot systems applicable for collaborative work towards common goals in manufacturing processes

在制造自动化中，对能够在工业制造过程中协作实现共同目标的多机器人应用的需求正在出现。这些应用程序在减少占空比、提高生产率和节省大量成本方面的优势是显而易见的。然而，由于复杂和耗时的编程过程，多机器人安装用于生产目的的适用性似乎受到限制，而且还由于缺乏开放的控制系统，使甚至异构机器人能够在协作环境中作为一个团队进行合作。为了克服这些缺陷，本项目将重点放在研究和技术开发上，使机器人团队能够自主规划和协调工作，并确保机器人的运动不会发生任何碰撞。应该避免显式编程。为了确保机器人团队与来自不同供应商的系统的应用，特别关注开发一个开放的控制概念，并使用适当的基于soa的基础设施和公共接口标准，以实现通信和交互中的互操作性。核心要素是计算机平台。它应集成所有必要的功能，以XML等元格式从制造过程的明确描述中提取相关数据，并自动规划、协调和控制机器人工作。在这种情况下，将特别考虑规划、优化和控制算法的发展，这些算法适用于在制造过程中由于工件重新定位的需求而动态变化的工作环境。研究和技术开发的预期结果应适用于由标准工业机器人组成的团队的协作自动化，也适用于具有冗余运动结构的系统。为了证明结果的技术可行性，将以焊接自动化领域的用例为例进行选择，因为该工业部门对具有先进规划和控制能力的多机器人装置的需求不断增加。

In manufacturing automation there is an emerging demand for multirobot applications which are capable of collaborating towards common goals in industrial manufacturing processes. The advantages of those applications are obvious in terms of reduced duty cycles, increased productivity, and considerable savings of costs. However the applicability of multirobot installations for production purpose seems to be restricted because of complex and time consuming programming procedures, but also due to the lack of open control systems which enable even heterogeneous robots to cooperate as a team in collaborative environments.To overcome these deficits, this project is focusing on research and technological development to enable robot teams to plan and coordinate the work autonomously and to assure robot movement free of any collision. Explicit programming should be avoided. In order to assure also the application of robot teams with systems from various suppliers, special focus is given to the development of an open control concept with an appropriate SoA-based infrastructure and common interface standards to enable interoperability in communication and interaction. Core element is a computer platform. It shall integrate all functions necessary to extract relevant data from an explicit description of the manufacturing process in a meta-format like XML, and to plan, coordinate, and control the robot work automatically. In this context special consideration will be given to the development of planning , optimization and control algorithms that are applicable to working environments that will change dynamically due to demands for workpiece reorientation during the manufacturing process. The results expected from research and technological development should be applicable for collaborative automation with teams built from standard industrial robots but also for systems with redundant kinematical structures. To prove the technical feasibility of the results, use-cases from the welding automation domain will be selected exemplarily because of the increasing demand for multirobot installations with advanced planning and control capabilities in this industrial sector.