To address the dynamic obstacle environment collision avoidance challenge of the marine autonomous surface ships (MASS), a decision-making method based on the deepq-learning (DqN) and velocity obstacle (VO) algorithm...
详细信息
To address the dynamic obstacle environment collision avoidance challenge of the marine autonomous surface ships (MASS), a decision-making method based on the deepq-learning (DqN) and velocity obstacle (VO) algorithm is proposed. Firstly, the encounter situation identification criteria are optimized, and a method for random collision scenario generation is designed. The model's performance is comprehensively evaluated by generating a wide variety of random collision scenarios which provide a broader assessment compared to manually set scenarios. Furthermore, a complete reward function for the dynamic collision avoidance problem is proposed, in which combines ship collision risk, the velocity obstacle method, and the International Regulations for Preventing Collisions at Sea (COLREGs). The MASS is not only guided towards the target by this reward function but is also ensured to comply with COLREGs during the collision avoidance process. It is worth noting that the trained model does not require retraining when faced with different numbers of target ships (TS). Simulation experiments are conducted with the trained model, involving random encounters with 1 to 10 TS in open waters. The results indicate that the proposed method demonstrates better collision avoidance performance compared to the DqN and proximal policy optimization algorithms.
The scarcity of space resources and backward scheduling management in block assembly yards have become constraints on shipyard logistics owing to demands in shipbuilding technology and scheduling efficiency. This arti...
详细信息
The scarcity of space resources and backward scheduling management in block assembly yards have become constraints on shipyard logistics owing to demands in shipbuilding technology and scheduling efficiency. This article proposes a novel integrated dynamic scheduling method for shipyard blocks considering adaptive adjustment in stockyard layout. First, an optimization model is established to minimize the interference block moves. Then, dynamic adaptive layout prioritization rules and multi-strategy block movement policies are introduced for different cases. A heap field dynamic layout method based on a binary search tree and deep reinforcement learning algorithms is integrated to seek optimal scheduling. The proposed algorithm is validated with 50 sets of actual block data, four road types and four stockyard types. The results indicate that the proposed method effectively utilizes stockyard layout while reducing interference blocks during the movement process by 56%. A stockyard aspect ratio of approximately 3.2 yields optimal block scheduling effectiveness.
In unmanned aerial vehicle ad-hoc network (UANET), the network topology changes with time due to the movement of the unmanned aerial vehicles (UAVs), which brings great challenges to the design of the routing protocol...
详细信息
In unmanned aerial vehicle ad-hoc network (UANET), the network topology changes with time due to the movement of the unmanned aerial vehicles (UAVs), which brings great challenges to the design of the routing protocol. In traditional routing protocols, when UANET topology changes, nodes cannot dynamically update neighbour nodes and topology information, and routing table calculation cannot accurately reflect the actual transmission path. As a result, network cannot meet the quality of service (qoS) requirements such as low end-to-end delay, high throughput and low packet loss rate. This paper proposes a dynamically optimized link state routing (OLSR) protocol based on deep q-network algorithm (DqN-OLSR). In this protocol, each node first adjusts the sending interval of Hello messages adaptively in real time, according to the position and speed information of its neighbour nodes. Then the protocol uses the DqN algorithm to dynamically adjust the flooding interval of topology control (TC) messages to improve the routing update capability of nodes. The simulation verifies that the UANETs under this protocol have higher throughput and less packet loss rate than ad-hoc on-demand distance vector (AODV), grid routing protocol (GRP) and OLSR protocols, at different movement speeds in random waypoint (RWP) and random walk mobile models. Under nomadic as well as pursue mobile models, DqN-OLSR performs consistently with OLSR qoS performance, with the best performance among all four protocols. By further adding positioning errors to the nodes, it shows that the proposed protocol has good robustness, and the qoS performance degradation keeps within a low level.
Allocating defense resources to specific lines can enhance the resilience of power systems against external damages. Considering the impact of information systems, a defense resource allocation model for cyber-physica...
详细信息
ISBN:
(纸本)9798350311259
Allocating defense resources to specific lines can enhance the resilience of power systems against external damages. Considering the impact of information systems, a defense resource allocation model for cyber-physical power systems (CPPS) is developed with the length of power lines as the defense cost. It is assumed that defense resources can reduce the probability of successful attacks. For this nonlinear programming (NLP) problem, an optimization-seeking method based on the deepq-network (DqN) algorithm is proposed. The model and algorithm are evaluated based on the IEEE-39 bus system. The results show that for small action sets, the method is in general agreement with the results obtained by the optimization solver BONMIN. In addition, the allocation strategies with different scales of resources and action sets are analyzed. These studies can provide ideas for the application of deep reinforcement learning (DRL) in resource allocation for power systems.
暂无评论