This paper provides an update to our ongoing Resilient Autonomous Systems (RAS) project that is being conducted at Embry-Riddle Aeronautical University and is supported by the Air Force Research Lab (AFRL). The objective of the RAS project is to develop a distributed multi-layer autonomous UAS planning and control technology for gathering intelligence in Anti-Access Area Denial (A2/AD) environments populated by intelligent adaptive adversaries. These resilient autonomous systems are able to navigate through hostile environments while collecting intelligence and minimizing the loss of assets. Our approach incorporates artificial life concepts, with the high-level architecture divided into three biologically inspired meta-layers: cyber-physical, reactive, and deliberative. Each meta-layer is a combination of N layers, each representing simple behavior, and collectively having a dynamic level of influence over the behavior of the agent. Algorithms within each layer act on a combination of perceptions, memory, and intents. Perceptions represent a filtered view of reality, abstracted in the layer immediately below, modified in the current layer, and passed to the next highest layer. Intents represent a recommended action output from the layer above, modified in the current layer, and passed to the next lowest layer. Each cycle, perceptions and intents are cleared. Memory permeates each layer, and allows the layers to retain some knowledge from previous cycles. Fast-reactive control systems in low meta-layers ensure a stable environment supporting cognitive function on higher meta-layers. The key concepts of our approach are: A layered architecture of intelligence on a spectrum from reactive to deliberative; the use of artificial life algorithms providing robust, complex behavior from a set of simple rules; communication of information between agents via observation; the use of a genetic algorithm (GA) to assist the team in selecting and adapting the best solutions for each layer. This distributed cooperative system of intelligent assets will provide adaptable, scalable performance to accomplish mission goals in challenging environments.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Highly Flexible Swarm Intelligence Algorithm for Resilient Autonomous Systems


    Contributors:


    Publication date :

    2018-09-01


    Size :

    1601635 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Multi-layer model of swarm intelligence for resilient autonomous systems

    Clifford, J. / Garfield, K. / Towhidnejad, M. et al. | IEEE | 2017


    Autonomous Car With Swarm Intelligence

    Anand, Y / Ajithkumar, Rahul | IEEE | 2019



    Swarm Intelligence Based Algorithm for Management of Autonomous Vehicles on Arterials

    Li, Lin / Hao, Ruochen / Ma, Wanjing et al. | British Library Conference Proceedings | 2018


    Swarm Intelligence Based Algorithm for Management of Autonomous Vehicles on Arterials

    Diao, Chenxue / Li, Lin / Hao, Ruochen et al. | SAE Technical Papers | 2018