Zichen Tang, Haihong E, Ziyan Ma, Haoyang He, Jiacheng Liu, Zhongjun Yang, Zihua Rong, Rongjin Li, Kun Ji, Qing Huang, Xinyang Hu, Yang Liu, Qianhe Zheng. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

This paper addresses the problem of optimal fault-tolerant attitude control with prescribed performance for a tailless flying wing aircraft under actuator faults and saturation. Firstly, by using the time-scale separation principle, the considered state variables of the aircraft are divided into two groups, the angular rate loop and the attitude loop, with different response speeds. Secondly, after prescribed performance bound (PPB) transformation, a novel error tracking nominal system is obtained, based on which an incremental neural network observer is designed to approximate the actuator faults. Thirdly, a functional performance index in terms of the actuator saturation is designed for the nominal system and the corresponding Hamilton-Jacobi-Bellman (HJB) equation is derived. A new single-network incremental adaptive dynamic programming (SIADP) algorithm is proposed to solve the optimal control problem. Moreover, the uniformly ultimately bounded stability of the closed-loop incremental system is proved by using the Lyapunov theory. Finally, simulations are given to illustrate the effectiveness of the proposed control approach.Note to Practitioners—The aim of this paper is to develop a new control scheme to realize active fault-tolerant control for tailless flying wing aircraft with actuator faults and saturation. The absence of vertical and horizontal tail fins results in a short longitudinal control moment arm, insufficient longitudinal control efficiency, deep coupling between longitudinal and lateral dynamics and poor stability of the tailless flying wing aircraft. Fault tolerance ability is necessary in flight control design, and the optimal and transient performance with control surface saturation should be considered in control practice. To solve these problems, a PPB-SIADP scheme combined with incremental adaptive observer approach is proposed, the tracking error can converge in an optimal way with prescribed performance. The proposed approach in this paper can be used as a valuable scheme for the tailless flying wing aircraft control law design in engineering practice.

Cybersecurity is of vital importance to industrial control systems (ICSs), such as ship automation, manufacturing, building, and energy automation systems. Many control applications require hard real-time channels, where the delay and jitter are in the levels of milliseconds or less. To the best of our knowledge, no encryption algorithm is fast enough for hard real-time channels of existing industrial fieldbuses and, therefore, made mission-critical applications vulnerable to cyberattacks, e.g., delay and data injection attacks. In this article, we propose a novel recursive watermark (RWM) algorithm for hard real-time control system data integrity validation. Using a watermark key, a transmitter applies watermark noise to hard real-time signals and sends through the unencrypted hard real-time channel. The same key is transferred to the receiver by the encrypted nonreal-time channel. With the same key, the receiver can detect if the data have been modified by the attackers and take action to prevent catastrophic damages. We provide analysis and methods to design proper watermark keys to ensure reliable attack detection. We use a ship propulsion control system for the simulation-based case study, where our algorithm smoothly shuts down the system after attacks. We also evaluated the algorithm speed on a Siemens S7-1500 programmable logic controller (PLC). This hardware experiment demonstrated that the RWM algorithm takes about 2.8 μs to add or validate the watermark noise on one sample data point. As a comparison, common cryptic hashing algorithms can hardly process a small data set under 100 ms. The proposed RWM is about 32 to 1375 times faster than the standard approaches.

Tidal bore is a peculiar nature phenomenon which is caused by the lunar and solar gravitation. Based on the physical characters of tidal bores, in this paper we propose a novel method to model and render this phenomenon, especially the tidal waves in Qiantang estuary. According to Boltzmann equation for tidal waves, we solve it with the novel triangle mesh of kinectic flux vector splitting (KFVS) mode. Then a method combining a curve forecasting wave and particles model is proposed to render the dynamic scenes of overturning tidal waves. Finally, with some rendering technologies, various realistic tidal waves under diversified conditions is rendered in real time.

This paper presents the development of a novel real-time operating environment for networked control systems (NCSs). An open-loop unstable magnetic-levitation (maglev) test bed was constructed and used to develop an NCS with real-time application interface (RTAI). A client-server architecture on a local-area network was developed with the network communication based on the user datagram protocol. The control loop of our NCS is closed over the network. This NCS structure gives the best flexibility and has significant economical merits. The implementation of an event-driven server and a time-driven client presented in this paper facilitates a simple timing scheme that does not require clock synchronization between the client and the server. A novel prediction scheme with a multiple-step-ahead control-signal generator is used to maintain system stability in the presence of excessive time delays and data-packet losses in the communication network. The performance of this NCS, based on the predictor algorithm, is demonstrated experimentally. The current system can compensate for up to 20% data-packet losses without losing stability with the maglev real-time-control test bed in the communication network. Our real-time operating environment also improved the command-following capability by a factor of 4 in terms of command frequency. Note to Practitioners- With the advancement in the automation industry, the need to perform complex remote operations has grown. Ever-increasing computational capabilities and advancements in the networking technology have aided researchers to develop architectures to implement control from a distance. In large-scale control applications of the modern industry, the functional agents, such as sensors, actuators, and controllers are geographically distributed. For smooth working of a control application, all of the agents have to exchange information through communication media. The results of this paper can help to design practical real-time networked control systems in modern industry.