Autonomous Assembly of Solar Array Modules by a Team of Robots

This project will investigate the hypothesis that Intelligent Precision Jigging Robots (IPJRs) and auxiliary robotic manipulators can autonomously perform the local level positioning and alignment required to enable precision assembly, disassembly, and reassembly of modular components. Key IPJR capabilities to be developed include: 1) high precision sensing and actuation; 2) state estimation using communication and simultaneous localization and mapping (SLAM); 3) robust failure detection and correction algorithms; and 4) high level artificial intelligence techniques in conjunction with low level control methods. This proposal will investigate the assembly of solar array modules to a backbone truss. The general process will use up to 6 IPJRs to grapple, precisely precision, and align a solar array module relative to a backbone truss and hold the module in place while the Lightweight Surface Manipulation System (LSMS) completes joining operations. FY16 Hardware benchmark will be a teleoperated assembly experiment that does not include simulated errors, disassembly, or reassembly. It will incorporate a SLAM algorithm for state estimation, but use high level commands given by human operators. Software benchmark will be a simulated assembly test with the following additional capabilities: disassembly by replacing step 4 with a bond cutting step, reassembly by reusing a previously disassembled solar array module, and error handling in spite of a variety of possible failures at each assembly step (such as IPJR grasp failure, IPJR deactivation, detachment of a solar panel, and failed sensors). Success will be defined by the fraction of completed trials. More »