Today's robots are good at executing programmed motions, but they do not understand their actions in the sense that they could automatically generalize them to novel situations or recover from failures. IMAGINE seeks to enable robots to understand the structure of their environment and how it is affected by its actions. "Understanding" here means the ability of the robot (a) to determine the applicability of an action along with parameters to achieve the desired effect, and (b) to discern to what extent an action succeeded, and to infer possible causes of failure and generate recovery actions. The core functional element is a generative model based on an association engine and a physics simulator. "Understanding" is given by the robot's ability to predict the effects of its actions, before and during their execution. This allows the robot to choose actions and parameters based on their simulated performance, and to monitor their progress by comparing observed to simulated behavior. This scientific objective is pursued in the context of recycling of electromechanical appliances. Current recycling practices do not automate disassembly, which exposes humans to hazardous materials, encourages illegal disposal, and creates significant threats to environment and health, often in third countries. IMAGINE will develop a TRL-5 prototype that can autonomously disassemble prototypical classes of devices, generate and execute disassembly actions for unseen instances of similar devices, and recover from certain failures. For robotic disassembly, IMAGINE will develop a multi-functional gripper capable of multiple types of manipulation without tool changes. IMAGINE raises the ability level of robotic systems in core areas of the work programme, including adaptability, manipulation, perception, decisional autonomy, and cognitive ability. Since only one-third of EU e-waste is currently recovered, IMAGINE addresses an area of high economical and ecological impact.

Currently industrial robots perform rigidly programmed tasks in highly-constrained settings. Any change in product or process requires costly restructuring of hardware and software. ReconCycle will address these issues by introducing the concept of robotic self-reconfiguration in the largely unconstrained domain of electronic waste recycling, which is still dominated by manual labor. Automation in this sector can benefit from the fact that very large batches of the same device type are to be processed but with some model differences and showing different states of damage. To be able to deal with each of these individual models, the robotic system requires flexible adaptation. Thus, the scientific objective of ReconCycle is to introduce in this sector self-reconfigurable hardware and software based on a reconfigurable robotic cell developed in a previous project. A two-step procedure is foreseen: When changing from one device-type to another, reconfiguration shall be performed in an interactive mode, where the application-engineer will be able to provide input. But, when changing from one device-model to another within a given device-type, the cell shall perform re-configuration on its own through a combination of sensorimotor learning approaches and other AI techniques. This constitutes the main novel scientific contributions of ReconCycle and is aimed at advancing our understanding of robotic perception-action processes in unconstrained industrial settings. We will use highly compliant soft robots and end-effectors, allowing humans to operate together with the machines to complete any missing steps. This reduces automation complexity and brings this project into a feasible regime for up to TRL 6. Electronic waste recycling is an important and strongly growing economic and environmental sector. The industrial objective of ReconCycle is to introduce here a much-increased level of automation resulting in a potentially high long-term impact on industry and society.

Specific raw materials become increasingly important to manufacture high level industrial products. Especially electronic equipment contains precious metals and a series of strategic raw materials. To date the material specific recycling is focused on mass stream concepts such as shredder processes and metallurgy to extract the high-value metallic constituents, i.e. copper, gold, silver. However, a series of critical elements cannot be recovered efficiently or is even lost in dust or residual fractions. The goal of ADIR is to demonstrate the feasibility of a key technology for next generation urban mining. An automated disassembly of electronic equipment will be worked out to separate and recover valuable materials. The concept is based on image processing, robotic handling, pulsed power technology, 3D laser measurement, real-time laser material identification (to detect materials), laser processing (to access components, to selectively unsolder these; to cut off parts of a printed circuit board), and automatic separation into different sorting fractions. A machine concept will be worked out being capable to selectively disassemble printed circuit boards and mobile phones with short cycle times to gain sorting fractions containing high amounts of valuable materials. Examples are those materials with high economic importance and significant supply risk such as tantalum, rare earth elements, germanium, cobalt, palladium, gallium and tungsten. A demonstrator will be developed and evaluated in field tests at a recycling company. The obtained sorting fractions will be studied with respect to their further processing and recovery potential for raw materials. Refining companies will define requirements and test the processing of sorting fractions with specific material enrichments. An advisory board will be established incorporating three telecommunication enterprises.