针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,...针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,LSTM)神经网络时延预测模型,在Matlab软件中搭建了该模型,并通过网络时延训练集数据对模型进行了训练,在网络时延测试集数据上对训练好的模型进行了测试,最后,就R2、RMSE和MAE评价指标对测试效果和GRU、BO-GRU以及LSTM预测方法进行了对比,结果表明,BO算法优化的LSTM神经网络时延预测模型RMSE和MAE评价结果更低,预测精度更高,预测效果更好,验证了提出的网络时延预测模型的有效性。展开更多
The path following problem for an underactuated unmanned surface vehicle(USV) in the Serret-Frenet frame is addressed. The control system takes account of the uncertain influence induced by model perturbation, externa...The path following problem for an underactuated unmanned surface vehicle(USV) in the Serret-Frenet frame is addressed. The control system takes account of the uncertain influence induced by model perturbation, external disturbance, etc. By introducing the Serret-Frenet frame and global coordinate transformation, the control problem of underactuated system(a nonlinear system with single-input and ternate-output) is transformed into the control problem of actuated system(a single-input and single-output nonlinear system), which simplifies the controller design. A backstepping adaptive sliding mode controller(BADSMC)is proposed based on backstepping design technique, adaptive method and theory of dynamic slide model control(DSMC). Then, it is proven that the state of closed loop system is globally stabilized to the desired configuration with the proposed controller. Simulation results are presented to illustrate the effectiveness of the proposed controller.展开更多
文摘针对通过无线通信网络实现远程控制的地面无人系统,分析了地面无人系统在工作的过程中,网络时延对系统的影响。基于网络时延的分布特性,提出了一种贝叶斯算法(Bayesian algorithm,BO)优化的长短期记忆(long-term and short-term memory,LSTM)神经网络时延预测模型,在Matlab软件中搭建了该模型,并通过网络时延训练集数据对模型进行了训练,在网络时延测试集数据上对训练好的模型进行了测试,最后,就R2、RMSE和MAE评价指标对测试效果和GRU、BO-GRU以及LSTM预测方法进行了对比,结果表明,BO算法优化的LSTM神经网络时延预测模型RMSE和MAE评价结果更低,预测精度更高,预测效果更好,验证了提出的网络时延预测模型的有效性。
基金Project(51409061)supported by the National Natural Science Foundation of ChinaProject(2013M540271)supported by China Postdoctoral Science Foundation+1 种基金Project(LBH-Z13055)supported by Heilongjiang Postdoctoral Financial Assistance,ChinaProject(HEUCFD1403)supported by Basic Research Foundation of Central Universities,China
文摘The path following problem for an underactuated unmanned surface vehicle(USV) in the Serret-Frenet frame is addressed. The control system takes account of the uncertain influence induced by model perturbation, external disturbance, etc. By introducing the Serret-Frenet frame and global coordinate transformation, the control problem of underactuated system(a nonlinear system with single-input and ternate-output) is transformed into the control problem of actuated system(a single-input and single-output nonlinear system), which simplifies the controller design. A backstepping adaptive sliding mode controller(BADSMC)is proposed based on backstepping design technique, adaptive method and theory of dynamic slide model control(DSMC). Then, it is proven that the state of closed loop system is globally stabilized to the desired configuration with the proposed controller. Simulation results are presented to illustrate the effectiveness of the proposed controller.
