Yi Ding

Trajectory planning plays a crucial role in ensuring the safe navigation of ships, as it involves complex decision making influenced by various factors. This paper presents a heuristic algorithm, named the Markov decision process Heuristic Algorithm (MHA), for time-optimized avoidance of Unmanned Surface Vehicles (USVs) based on a Risk-Sensitive Markov decision process model. The proposed method utilizes the Risk-Sensitive Markov decision process model to generate a set of states within the USV collision avoidance search space. These states are determined based on the reachable locations and directions considering the time cost associated with the set of actions. By incorporating an enhanced reward function and a constraint time-dependent cost function, the USV can effectively plan practical motion paths that align with its actual time constraints. Experimental results demonstrate that the MHA algorithm enables decision makers to evaluate the trade-off between the budget and the probability of achieving the goal within the given budget. Moreover, the local stochastic optimization criterion assists the agent in selecting collision avoidance paths without significantly increasing the risk of collision.

Ship collision avoidance is a complex process that is influenced by numerous factors. In this study, we propose a novel method called the Optimal Collision Avoidance Point (OCAP) for unmanned surface vehicles (USVs) to determine when to take appropriate actions to avoid collisions. The approach combines a model that accounts for the two degrees of freedom in USV dynamics with a velocity obstacle method for obstacle detection and avoidance. The method calculates the change in the USV’s navigation state based on the critical condition of collision avoidance. First, the coordinates of the optimal collision avoidance point in the current ship encounter state are calculated based on the relative velocities and kinematic parameters of the USV and obstacles. Then, the increments of the vessel’s linear velocity and heading angle that can reach the optimal collision avoidance point are set as a constraint for dynamic window sampling. Finally, the algorithm evaluates the probabilities of collision hazards for trajectories that satisfy the critical condition and uses the resulting collision avoidance probability value as a criterion for course assessment. The resulting collision avoidance algorithm is optimized for USV maneuverability and is capable of handling multiple moving obstacles in real-time. Experimental results show that the OCAP algorithm has higher and more robust path-finding efficiency than the other two algorithms when the dynamic obstacle density is higher.

Adaptive boost (AdaBoost) is a prominent example of an ensemble learning algorithm that combines weak classifiers into strong classifiers through weighted majority voting rules. AdaBoost’s weak classifier, with threshold classification, tries to find the best threshold in one of the data dimensions, dividing the data into two categories-1 and 1. However, in some cases, this Weak Learning algorithm is not accurate enough, showing poor generalization performance and a tendency to over-fit. To solve these challenges, we first propose a new Weak Learning algorithm that classifies examples based on multiple thresholds, rather than only one, to improve its accuracy. Second, in this paper, we make changes to the weight allocation scheme of the Weak Learning algorithm based on the AdaBoost algorithm to use potential values of other dimensions in the classification process, while the theoretical identification is provided to show its generality. Finally, comparative experiments between the two algorithms on 18 datasets on UCI show that our improved AdaBoost algorithm has a better generalization effect in the test set during the training iteration.

In order to ensure efficiency of Trans-Asia Gas Pipeline, engineers in Sino-Petroleum International Company optimized pipeline operation from a whole view through simulated optimization and unit control optimization. The network model of ABC line is established using TGNET, and the error of model is adjusted to the acceptable range by using the practical data. Flow distribution of AB line and C line is optimized to ensure the minimum fuel cost. Principle of gas pack control is built in different working conditions. And the compressor unit is controlled according to the simulation results. Performance of compressor unit is monitored and control mode is regulated to a better way on the basis of data analysis and investigation. After more than one year of exploration and practice, the operation optimization methodology of the Trans-Asia Gas Pipeline has been established. Energy-saving effect is remarkable.