AvantGuard is an instrument with which researchers can study the interplay of human direction and autonomous behavior in conducting a simulated mission where a human supervises multiple autonomous Unmanned Air Vehicles (UAV). AvantGuard is a computer game in which UAVs are used to protect a convoy moving through unsecured urban terrain under conditions of asymmetric conflict. Adversaries hide among the residents and prepare an ambush. The operator directs small UAVs and studies the resulting sensor stream to find the ambush before the convoy arrives in the trap. Autonomy is decomposed into four cognitive phases: look. see, decide, act. Each is matched to an exemplary task: sensor location, image analysis, threat assessment, and convoy rerouting. Each of these is independently assigned a Level of autonomy, or an adaptive strategy. Using sophisticated game techniques, a high-performance UAV simulation game prototype was developed , as well as the design for the experimenter interface.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    Machine Learning and Instrument Autonomy

    Bue, Brian / Estlin, Tara / Altinok, Alphan et al. | NTRS | 2016




    SPACECRAFT AUTONOMY ISSUES: PRESENT AND FUTURE

    Horvath, J.C. | Online Contents | 1996


    Autonomy issues for an operational space station

    Daly, K. C. / Cox, K. J. | NTRS | 1983