AN EVOLUTIONARY ALGORITHM FOR NEAR-OPTIMAL AUTONOMOUS RESOURCE MANAGEMENT
Andrew Carrel, Phil L. Palmer · 2005
A new autonomy algorithm, described in this paper, has been shown to produce significant improvements in the rate of execution of spacecraft operations in a real scenario. This algorithm is able to manage a number of generic resources autonomously and schedules generic operations, each consisting of one or more related tasks. This Near-optimal Evolutionary Autonomous Task-manager (NEAT) uses a genetic algorithm to maximise the throughput of the system, accounting for priorities where appropriate. Since onboard computing power is limited NEAT always has a valid operations schedule available during the evolution process such that a decision can be made at any time. NEAT has been applied to the case of SSTL’s UK-DMC