A preliminary study of an intent-recognition-based traded control architecture for high latency telemanipulation

摘要: Despite the availability of increasingly more robust machine perception and control algorithms, human operators are still needed to directly robotic systems in extreme environments where there is a high cost failure. This paper reports novel architecture for semi-autonomous teleoperation over high-latency telemetry results 15-subject user study used evaluate that architecture. The intent-recognition-based traded (IRTC) recognizes operator intent based on remote scene task modeling, automatically trades an autonomous reactive subsystem. Each subject participated one two-hour session during which they controlled single simulated redundant manipulator with 4000ms closed-loop latency. Subjects continuously assemble disassemble 3D lattice structure or without IRTC-based assistance. showed IRTC system improved performance multi-object assembly task. All source code conducting experiment including simulation data processing available online under permissive open-source license.