Innovative adaptive autopilot design for uninhabited surface vehicles
Innovative adaptive autopilot design for uninhabited surface vehicles
复制标题
DOI：
10.1049/cp.2014.0677
复制
发表时间：
2014-09
期刊：
影响因子：
2.9
通讯作者：
A. Annamalai;R. Sutton;Chenguang Yang;P. Culverhouse;Sanjay K. Sharma
中科院分区：
文献类型：
--
作者：
A. Annamalai;R. Sutton;Chenguang Yang;P. Culverhouse;Sanjay K. Sharma
在现实生活中，动力的突然变化经常会导致任务中止，导致它必须在附近的其他海洋船只受到损害之前获救。本文描述的一种创新的自适应自动驾驶仪设计已经适当地解决了这个问题。基于加权最小二乘的模型预测控制具有在线自适应特性。即使是随机初始化，WLS通过保持系统参数的间歇连续值和周期性地重新初始化协方差矩阵，也取得了显著的改进。此外，25秒的时间范围似乎是重新初始化参数的最佳时间。这种新颖的方法使自动驾驶仪能够很好地应对系统动力学的重大变化，并使usv能够完成预期的任务。
In real life situations, very often a sudden change in dynamics results in missions being aborted which results in it having to be rescued before damage to other marine craft in the vicinity is caused. This problem has been suitably dealt with by an innovative adaptive autopilot design that is described herein. A model predictive control (MPC) adopts an online adaptive nature by utilizing weighted least squares (WLS). Even with random initialization, significant improvements were achieved by WLS by maintaining the intermittent continuous values of system parameters and periodically reinitializing the covariance matrix. Also, a time frame of 25 seconds appears to be the optimum to reinitialize the parameters. This novel approach enables the autopilot to cope well with significant changes in the system dynamics and empowers USVs to accomplish their desired missions.