基于改进粒子群算法的USV航向分数阶控制

针对欠驱动水面船舶（underactuated surface vessel，USV）航向保持稳定性问题，对船舶自动舵控制系统设计了分数阶PIλDμ控制器。积分阶次λ和微分阶次μ的引入使得分数阶比例积分微分（proportion integration differentiation, PID）PIλDμ控制器具有更好的鲁棒性和抗扰动能力，但同时也加大了算法设计的难度。使用改进粒子群算法对分数阶PIλDμ控制器参数进行整定，即解决了粒子群算法容易使粒子陷入局部最优问题，又解决了分数阶PIλDμ控制器整定参数多、设计复杂问题。通过仿真对比实验，结果表明，该控制器能很好地根据船舶动态特性变化，自动进行适应性参数优化，具有跟踪速度快、航向控制超调小以及抗扰性强等优点。

Fractional order control of USV course based on improved PSO algorithm

To improve the course stability of underactuated surface vessel (USV), a fractional order PIλDμ （proportion integration differentiation, PID）controller is applied to ship autopilot, which is more flexible, robust and has stronger disturbance rejection ability with the integral order λ and differential order μ. However， difficulties for optimization also increase. A new method of designing PIλDμ controller based on the improved particle swarm optimization (IPSO) algorithm is proposed, which solves the problems of the particles falling into local optimum and the design complexity in fractional order PIλDμ controller design. The fractional order PIλDμ controller based on IPSO is compared with the traditional PSO PID controller under the same conditions. The simulation results show that，the controller can optimize the adaptive parameters well and automatically, and has a high tracking speed, small overshoot and strong immunity.