Logic Control and “Reactive” Systems: Algorithmization and Programming
Automation and Remote Control
Learning Deterministic Finite Automata with a Smart State Labeling Evolutionary Algorithm
IEEE Transactions on Pattern Analysis and Machine Intelligence
Tools for support of automata-based programming
Programming and Computing Software
Q&A: Talking model-checking technology
Communications of the ACM - Web science
Evolving finite state transducers: some initial explorations
EuroGP'03 Proceedings of the 6th European conference on Genetic programming
Genetic programming with fitness based on model checking
EuroGP'07 Proceedings of the 10th European conference on Genetic programming
Proceedings of the 14th annual conference companion on Genetic and evolutionary computation
Learning finite-state machines with ant colony optimization
ANTS'12 Proceedings of the 8th international conference on Swarm Intelligence
MuACOsm: a new mutation-based ant colony optimization algorithm for learning finite-state machines
Proceedings of the 15th annual conference on Genetic and evolutionary computation
An evolutionary methodology for automatic design of finite state machines
Proceedings of the 15th annual conference companion on Genetic and evolutionary computation
| Hi-index | 0.00 |
In this paper, we describe the method of finite state machine (FSM) induction using genetic algorithm with fitness function, cross-over and mutation based on testing and model checking. Input data for the genetic algorithm is a set of tests and a set of properties described using linear time logic. Each test consists of an input sequence of events and the corresponding output action sequence. In previous works testing and model checking were used separately in genetic algorithms. Usage of such an approach is limited because the behavior of system usually cannot be described by tests only. So, additional validation or verification is needed. Calculation of fitness function based only on verification do not perform well because there are very few possible values of fitness function (verification gives only "yes" or "no" answer). The approach described is tested on the problem of finite state machine induction for elevator doors controlling. Using tests only the genetic algorithm constructs the finite machine working improperly in some cases. Usage of verification allows to induct the correct finite state machine.