Xiumin Chu

Trajectory Tracking Control for Underactuated Surface Vessels Based on Nonlinear Model Predictive Control

An autonomous vessel can improve the intelligence, efficiency and economy of shipping logistics. To realize autonomous navigation control of underactuated vessels (USVs), this paper presents a controller that can make USV track a reference trajectory with only 2 inputs, i.e., surge and yaw. A nonlinear state-space model with 2 inputs considering environmental disturbances induced by wind, current and waves for a 3 degree of freedom (DOF) surface vessel is considered. Based on this model, a trajectory tracking controller using Nonlinear Model Predictive Control (NMPC) is designed. System constraints on inputs, input increment and output are incorporated in the NMPC framework. Simulation results show that the controller can track an ellipse trajectory well and that the tracking errors are within acceptable ranges, while system constraints are satisfied.

Haze removal for a single inland waterway image using sky segmentation and dark channel prior

Haze significantly degrades the visibility for ship navigation and traffic monitoring in Chinas inland waterways. In this study, the authors propose a novel haze-removal method based on sky segmentation and dark channel prior to restore images. Sky segmentation is accomplished by using robust image matting and region growth algorithms. Then, the average image intensity of the sky region is chosen as the atmospheric light value to address the defect of dark channel prior. Experimental results show that their method can restore inland waterway images effectively; restored images are more natural and smoother than those obtained by the state-of-the-art haze removal algorithms.

Simulation models of vessel traffic flow in inland multi-bridge waterway

In an inland multi-bridge waterway, vessel behaviors are influenced and restricted by various factors such as people, ships, environment and management etc.; the behaviors are always different with the temporal and spatial change; it is also difficult to use accurate mathematical model to describe the behaviors; and moreover water traffic is not suitable for experiments using real vessels. Hence people naturally think that the simulation is a good method to mimic behaviors of the vessels. This paper puts forward simulation models for simulating vessel traffic in the inland multi-bridge waterway. Main contents of this paper are as follows: probability distribution models of vessel traffic flow are obtained through the analysis of historical data of vessel traffic flow in the bridge area waterway; vessel generation model and route model is setup using the Monte Carlo method; the navigation behavior simulation model is present using the Multi-Agent method and the vessel agents are with the perception, interaction and reasoning ability, obey the traffic rules in different legs, and can finish the overtaking, following and obstacle avoidance behavior during the voyage. This research can quantitatively or visually analyzes behavior characteristics of the vessel, and offer means and methods for the evaluation of channel transit capacity, navigation risk, and organization and management ability.

The Error Distinguishing of Automatic Identification System Based on Improved Evidence Similarity

There is no fault-tolerant design in Automatic Identification System data link, so the fault message is inevitable. The DS theory is able to simulate the artificial judgment on AIS messages, but the motion and lane evidences conflict with each other occasionally. The improved evidence similarity method would solve the problem, simulate human thinking, take the evidence priorities as a new parameter to revise the combination, and improve the model. In field testing, the method shows very good performance.

A Quantificational Description Method of Vessel Track Based on AIS Data

The more accurate the vessel track can be described, the more information can be provided for analyzing the characteristic of vessel traffic, the path of the vessel, and the accidents on the water. Based on AIS information, a new method of describing the vessel track is designed. The line track when the vessel is on a straight course can be described by a line with a quantificational slope. The curve track when the vessel is in the beginning and ending turning process can be described by cubic curve with two quantificational boundary conditions. The curve track when the vessel is in steady circle can be described by an arc with quantificational center point, radius and course angle. For a case study, a serial AIS information is selected to describ the track using this method. The track is analyzed by more frequent GPS information. The results show that the track described by the method in this paper is more similar to the real track, and the errors are smaller, especially when the vessel is turning course.

Technology of Information Collection and Analysis about Steering Operation Behavior of Inland Waterway Sailing Ship

ABSTRACT He, H.; Xiong, J.; Liu, M.; Chu, X.; Liu, C.; Shi, L., and Wang, L., 2015. Technology of information collection and analysis about steering operation behavior of inland waterway sailing ship. Steering operation behavior is an important factor which affects the safety of ship navigation. This article used suitable method of information acquisition and information transmission and analyzed key technologies of steering behavior data collection, transmission and display, and then developed a monitoring system software for inland waterway ship sailing condition. This software is capable of tracking the ship in real time, displaying the static and dynamic information, operation information and environment information, as well as querying the historical trend functions. After the application, it shows that the software can display real-time steering operation behavior and realize remote management.

Method of Generating Simulation Vessel Traffic Flow in the Bridge Areas Waterway

Ship-ship or ship-bridge collision accidents are easy to occur when ships pass through the inland bridge areas waterway. Simulation based research on traffic flows in bridge areas waterway helps to reduce the accidents. In this paper, historical data about vessel traffic flow in bridge areas waterway has been collected and probability distributions of attributions of vessel traffic flows are obtained by analyzing the data. Then a route model of traffic flows and a motion model of vessels are presented. A generating algorithm of vessel traffic flows is also proposed and the algorithm is implemented using C++ programming language. Comparing the simulation data with actual data shows that the simulation modes and their algorithm are valid.

Marine diesel engine speed control based on adaptive state-compensate extended state observer-backstepping method

The marine diesel engine propulsion system is a nonlinear system with time delay. In order to realize the accurate and real-time control of the marine diesel engine speed, a new method based on state-compensate extended state observer, backstepping method and beetle antennae search algorithm, that is, adaptive state-compensate extended state observer-backstepping, is proposed. First of all, the response relationship model between the engine speed and the fuel injection is established on the basis of the mean value model of diesel engine. Then, to deal with the load disturbances and model parameter perturbation of diesel engine, a state-compensate extended state observer is used to estimate lumped disturbances and states of the diesel engine, and a backstepping method combined with the state-compensate extended state observer, namely state-compensate extended state observer-backstepping, is used to control the marine diesel engine speed. Then, an adaptive state-compensate extended state observer-backstepping controller is proposed by introducing the beetle antennae search algorithm for online optimization of the control parameters. Finally, simulation experiments based on the model of the 12K98ME marine diesel engine are conducted to verify the effectiveness of the proposed controller under conditions of random disturbances, sudden dumping load and parameter perturbation. The experiment results show that the proposed adaptive state-compensate extended state observer-backstepping control method has a better control effect and stronger disturbance rejection ability in comparison of the standard linear active disturbance rejection control.

A hybrid model for water level forecasting: A case study of Wuhan station

Water level forecasting has been recognized as a classic problem in the field of time series analysis and is closely related to the maritime administration. A hybrid model is thus presented for a case study of Wuhan station which locates at the middle reach of the Yangtze River. The daily water level of the Wuhan station is affected by the upstream runoff and human activities (waterway regulation projects, sand mining etc.), which lead to the complexity of its short-term forecasting. Time sequences of daily water level are derived from the Changjiang Waterway Bureau, on basis of which a hybrid ANN-Kalman filtering model is established. Its performance is evaluated and compared with single ANN model. The forecasting results are overall favorable. The hybrid model could be further integrated with the Yangtze River Electronic Channel Chart Display and Information System and provides an alternative way of daily water level predictions.

A variational approach for adaptive underwater sonar image denoising

Underwater sonar imaging is a significant means for a variety of maritime work. However, the sonar images are corrupted by signal dependent speckle noise, which restricts the potential practical applications. In this paper, we propose a novel variational approach that addresses the adaptive sonar image denoising problem. To accurately describe the signal dependent characteristic of noise for real sensors, we utilize a generalized noise model, which can be adapted to represent various types of noise. With the generalized noise model, we formulate the denoising problem via a variational approach. Finally, we present a modified primal and dual method to efficiently solve the variational minimization problem. Experimental results on both simulated and real data validate the effectiveness and efficiency of our method.