W. Liu,∗,∗∗ L. Gao,∗,∗∗ Y. Ding,∗ J. Xu,∗ and Y. Ji∗


Autonomous underwater vehicles, controller area network, experi- mental platform, scheduling, genetic algorithms, simulation


Autonomous underwater vehicles (AUVs) have many scientific, mil- itary and commercial applications. Typical AUV development re- quires critical system design and many costly underwater trials during which system specifications can be validated. Modelling and simulation provide a cost-effective method for carrying out pre- liminary component or system testing and verification. Controller area network (CAN) was originally developed by Bosch for use in automobiles has become an international standard. Although CAN was developed as a vehicle network standard, the majority of CAN applications exist outside of the automotive industry. One of the great advantages of CAN-bus architecture is the modularity. This paper introduces a PC-based CAN-bus experimental platform for AUV system simulation and performance evaluation. The implemen- tation and consideration on the hardware-in-the-loop experimental platform are presented. The genetic algorithm (GA) scheduling ap- proach is introduced in realizing CAN-bus real-time periodic message transmission. The mathematical models of the AUV are introduced in real-time mathematical simulation. The related timing analysis and simulation results are given and they showed the applicability and availability of the experimental platform.

Important Links:

Go Back