CASRO: a path optimization strategy for USV–UAV collaborative exploration of maritime target

Path planning is one of the fundamental issues in the research field of robot. In the last decades, Rapidly‐exploring Random Tree star (RRT*) was one of the chosen path planner for robot with its probabilistically complete. In this paper, an improved bidirectional RRT* path planner for robot is proposed (Bi‐RRT*), which has planned path with safety and smoothly guarantee. The main contributions of this paper include, firstly, combining the backtracking idea of the Quick‐RRT algorithm with the greedy search strategy of the RRT‐Connect algorithm, which can reduce the path cost and improve the operation efficiency of the algorithm. Second, in order to further reduce the running time of the algorithm, a new collision detection algorithm based on cross‐product is proposed to replace the conventional collision detection strategy. To ensure the quality of the path, two post‐processing strategies, path optimization strategy and path smoothing strategy, are proposed. Path optimization strategy is based on triangle inequality, and two optimization methods are proposed to effectively reduce the path cost. The path smoothing strategy based on the Bezier curve improves the continuity of a smooth arc and is better applied to path smoothing. Simulation results for both virtual and real environments show the advantages of combining Quick‐RRT and RRT‐Connect and verify the effectiveness of the proposed collision detection algorithm. Compared with similar algorithms, the proposed Bi‐RRT algorithm has higher operational efficiency, smaller path cost, smoother planned paths, and improved path quality. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

With the rapid development of the economy, the fourth party logistics (4PL) has become an indispensable part of economic development, and the path optimization is an important topic in the research of 4PL routing problem. This paper studies the fourth party logistics routing problem with random demand and uncertain transportation time and transfer time by considering the randomness and uncertainty of the environment. In order to solve the path optimization strategy that minimizes the total transportation cost, this paper establishes an uncertain random programming model under the constraints of the total transportation time, the carrying capacity of the logistics provider, the transfer capacity of the transfer station, and the demand of the distributor. Subsequently, this paper transforms the uncertain random programming model into an equivalent deterministic programming model based on the distribution functions of random demand and uncertain time, and verifies the effectiveness of the model through two examples. The results show that the uncertain random programming model can give a reasonable path optimization strategy.

A Topological Map-based Path Coordination and Optimization Strategy for Autonomous Parking

In this paper, we present a novel integrated vehicular system using collaborative unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) for autonomous exploration, mapping, and navigation in GPS-denied 3-dimensional (3-D) unknown environments. The system implements a novel two-layered exploration strategy to decompose the perception task into a coarse exploration layer and a fine mapping layer. The coarse exploration makes use of a UGV to carry out fast autonomous exploration and active 2.5D simultaneous localization and mapping (SLAM) to generate a coarse environment model, which serves as a navigation reference for subsequent complementary 3-D fine mapping conducted by a UAV. The two layers share a novel optimized exploration path planning and navigation framework, which provides optimal exploration paths and integrates the collaborative exploration and mapping efforts through an OctoMap-based volumetric motion planning interface. The proposed system provides an efficient pipeline of fast environment perception taking advantages of the agility of the UAVs as well as the powerful computation resource aboard UGVs, also allowing assistive local perception with augmented object information when necessary. The effectiveness of our system is verified by both simulations and experiments, which demonstrate its capability of implementing heterogeneous UAV and UGV collaborative exploration and structural reconstruction of the environments through active SLAM, providing optimized perception for navigation tasks.

In facility agriculture, automated transplanter is used to transplant the healthy seedlings in vacant cells to replace the bad ones. It can make full use of space and nutrition. And the working efficiency could be improved by selecting a shorter transplanting path length. In this paper, a path optimization strategy based on tray position control and greedy algorithm (TPCG) was introduced. 260 samples with randomly location vacant holes tray were used to do transplanting path planning simulation. And TPCG was compared with fixed sequence (FS). The results showed TPCG could shorten over 18% of the path length for 32 cell trays, and more than 16% of path length for 128 cell trays. The transplanting efficiency was improved and the operation time was 0.302s which could meet the real-time requirement of transplanting task.

To address the problem of identifying teaching behavior characteristics and optimizing the path for improving digital literacy abilities of university teachers, a multidimensional AI analysis framework integrating computer vision, semantic analysis, graph neural networks, and reinforcement learning is constructed. By collecting multimodal data and modeling graph structures, accurate qualitative and structural expression of teaching behavior characteristics is achieved. Combined with the five-dimensional competency system, a behavior driven evaluation model and path recommendation mechanism are established. Adopting the path optimization strategy of DQN, constructing an intelligent intervention tool and conducting experimental verification to verify the superiority of the model in terms of positioning accuracy, improvement of quality ability, and cross school transfer ability. The research provides intelligent solutions for the development of teachers' abilities and technical support for the digital transformation of the education system.

Data collection is the basic purpose of deploying in heterogeneous WSNs for Internet of things, and the problem of data collection is the key problem that needs to be solved in heterogeneous WSNs. How to collect energy-efficient and reliable data is one of the key technologies of heterogeneous WSNs. Collecting the sensor node data by mobile sink is an effective measure to solve data collection efficiency. To this end, a data collection strategy of mobile sink for heterogeneous WSNs based on pigeon-inspired optimization by PSO algorithm is proposed. The proposed algorithm uses the improved pigeon-inspired optimization by particle swarm optimization algorithm to select the best dwell point and then regards the construction of the moving path based on the dwell point as a traveling salesman problem to optimize the moving path and solve the optimal moving path. The experimental analysis and simulation results show that, compared with other algorithms, the algorithm proposed in this paper can effectively prolong the lifetime of the network and reduce the delay of data collection, increasing the amount of data collection.

With the rapid development of the logistics industry, unmanned logistics fleets play an increasingly important role in distributing goods in the region. In this paper, deep reinforcement learning algorithms are proposed, which are applied to the distribution path planning problem of unmanned logistics fleet with time window and the distribution path planning problem of unmanned logistics fleet with time window considering regional congestion. Based on deep reinforcement learning, an Actor policy network model incorporating the attention mechanism and a Critic value judgment network are designed. Through the parameterized probability estimator for model input state output node probability estimation, the reinforcement learning reward function, state transfer function, masking scheme, decoding strategy, objective and loss function, and the reinforcement learning algorithm that combines the ideas of AC and round updating used to train the policy network model and the value network model are given for the VRP problem with time window. Finally, an arithmetic example is used to explore the impact of different combinations of working hours and different vehicle loading capacities in a round on the total cost. Under three customer node sizes, the optimal solution occurs under the combination of 8 hours of working hours and 200 vehicle loading capacity, and the reasonable arrangement of different types of unmanned delivery fleets and working hours can effectively reduce the total cost in the delivery process.