Auction-based node selection of optimal and concurrent responses for a risk-aware robotic sensor network

• Main Authors: McCausland, Jamieson, Abielmona, Rami, Falcon, Rafael, Cretu, Ana-Maria, Petriu, Emil
• Format: Conference Proceeding
• Language: English
• Subjects: Biological cells; critical infrastructure protection; fuzzy logic; genetic algorithm; Measurement; Optimization; Protocols; Robot kinematics; Robot sensing systems; robotic sensor network; self-organization
• DOI: 10.1109/ROSE.2013.6698432

In this paper, an auction-based node selection technique is considered for a risk-aware Robotic Sensor Network (RSN) applied to Critical Infrastructure Protection (CIP). The goal of this risk-aware RSN is to maintain a secure perimeter around the CIP, which is best maintained by detecting high-risk network events and mitigate them through a response involving the most suitable robotic nodes. These robotic nodes can operate without the use of a centralized system and select amongst themselves the nodes with the best fitness to risk mitigation plan. The robot node that is first aware of a high-risk event becomes an auctioneer. The risk mitigation task is advertised to the entire network. Each robotic node is responsible for calculating their bid metric (i.e. availability metric) for the risk mitigation task. We employ fuzzy logic in the process of the bid calculation, which incorporates the battery level, distance to the event, and redundant coverage to produce an appropriate bid value. The auctioneer only considers the top bidders. The nature of this system is to permit simultaneous mitigation plans to execute on a single RSN by effectively segmenting the network into discrete autonomous groups. Each autonomous group will utilize an evolutionary multi-objective algorithm - the Non-Dominated Sorting Genetic Algorithm (NSGA-II) - to optimize the segment's topology to mitigate the risk. A chromosome length is determined by the number of bids received, but the NSGA-II explored to separate solution spaces to achieve optimal Pareto results. The NSGA-II will seek optimal node positions and determine the optimal set of robotic nodes to utilize of the bids received. The NSGA-II will produce a set of optimized responses for each network segment for a security operator to pick the most suitable response.