Task Allocation for Integrated Modular Avionics Using Particle Swarm Optimization

Recently the Integrated Modular Avionics (IMA) architecture becomes popular as an alternative to the traditional federated architecture. An IMA platform consists of multiple processing nodes interconnected via high-speed avionics data networks. A set of pre-defined tasks need to be assigned to different processing nodes with resource constraints before execution. This paper presents a task allocation algorithm for IMA systems based on the principles of particle swarm optimization (PSO). It tries to find the optimal task assignment with minimum communication costs and balanced work load. The algorithm guarantees that all the hard real-time tasks in the IMA platform can meet their deadlines, and the pre-allocated memory for hosted partitions will not exceed the capability of the processing node. Experimental results manifest the validity of the algorithm.