Over the past several years the Center for Evolutionary Computation and Automated Design at the Jet Propulsion Laboratory has developed a technique based on Evolutionary Computational Methods (ECM) that allows for the automated optimization of complex computationally modeled systems. An important application of this technique is for the identification of emergent behaviors in autonomous systems. Mobility platforms such as rovers or airborne vehicles are now being designed with autonomous mission controllers that can find trajectories over a solution space that is larger than can reasonably be tested. It is critical to identify control behaviors that are not predicted and can have surprising results (both good and bad). These emergent behaviors need to be identified, characterized and either incorporated into or isolated from the acceptable range of control characteristics. We use cluster analysis of automatically retrieved solutions to identify isolated populations of solutions with divergent behaviors.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Evolutionary Computation for the identification of emergent behavior in autonomous systems


    Contributors:


    Publication date :

    2009-03-01


    Size :

    2281065 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Adaptive Decentralized Emergent Behavior Based Autonomous Platooning

    Sreenivasamurthy, Shesha / Obraczka, Katia | IEEE | 2024



    Automatic Testing of an Autonomous Parking System Using Evolutionary Computation

    Buhler, O. / Wegener, J. / Society of Automotive Engineers | British Library Conference Proceedings | 2004


    Automatic Testing of an Autonomous Parking System Using Evolutionary Computation

    Wegener, Joachim / Bühler, Oliver | SAE Technical Papers | 2004