A learning-guided optimization approach, i.e., a $Q$-Learning-guided Memetic Algorithm (QLMA), is proposed to solve path planning problems in unknown environments. Traditional path planning algorithms in unknown environments have the disadvantages of low efficiency and stability problems. Integrating learning strategies into a memetic algorithm is a natural way to enhance search performance. The proposed algorithm can be divided into two stages: learning and optimization. A $Q$-learning algorithm implements the learning stage, while the optimization stage is implemented by hybridization of a Genetic Algorithm (GA) and Simulated Annealing (SA) algorithm, i.e., a memetic algorithm. The learning stage aims to acquire information from current search states. The $Q$-learning algorithm’s environmental learning capability is formed by rewards received from changes in distance to the target. Therefore, the path strategy could be adjusted continually, and effective paths could be identified. The optimization stage aims to utilize the obtained information and enhance the quality of known paths. A GA improves the global search ability, while an SA algorithm enhances the local search ability. The experimental study was conducted on 30 test scenarios with three types of index points. The proposed algorithm outperforms the other two algorithms significantly in the test scenarios.

As a practical solution that could reduce the communication and computation load of central servers in digital factories, edge computing has been widely used in modern industry. In mobile edge computing, a reasonable offloading strategy can balance the computation load and reduce the energy consumption of mobile devices, which is the key to optimizing network operation. In this paper, a Relative Position-based Bacterial Foraging Optimization algorithm with Dropout strategy, RPBFO-D, is proposed to optimize the computation offloading problem. A many-to-many relationship model of devices-tasks-servers is established, comprehensively considering the time delay and energy consumption, and RPBFO-D is proposed to solve the problem. In this algorithm, the structure and operators of the original BFO are redesigned, and the dropout strategy of the neural network maintains diversity. Experiments with parameter settings demonstrate the effectiveness of the dropout strategy. Results show that RPBFO-D has better convergence accuracy than comparison algorithms, which demonstrates that it is a competitive approach for computation offloading.

Most existing classification algorithms for cardiovascular disease are limited to specific diseases and cannot categorize the severity of the diseases. These algorithms still need to be improved in terms of accuracy and generalizability. Therefore, a hybrid Improved Brain Storm Optimization with Support Vector Machine (IBSO-SVM) for cardiovascular disease classification is proposed. In this study, a knowledge-driven intelligent initialization method is proposed to enhance the optimization capability of IBSO and the accuracy of IBSO-SVM. Experimental evaluations are conducted on multiple real-world datasets, and the results demonstrate the superior performance of IBSO-SVM in cardiac disease datasets. The accuracy of BSO-SVM reaches 100% on the Heart Failure and Heart Disease datasets, and the accuracy of IBSO-SVM reaches 99% on the Stroke dataset and 88% on the Cardiovascular disease dataset.

With continuous expansion of satellite applications, the requirements for satellite communication services, such as communication delay, transmission bandwidth, transmission power consumption, and communication coverage, are becoming higher. This paper first presents an overview of the current development status of Low Earth Orbit (LEO) satellite constellations, and then conducts a demand analysis for multi-satellite data transmission based on LEO satellite constellations. The problem is described, and the challenges and difficulties of the problem are analyzed accordingly. On this basis, a multi-satellite data-transmission mathematical model is then constructed. Combining classical heuristic allocating strategies on the features of the proposed model, with the reinforcement learning algorithm Deep Q-Network (DQN), a two-stage optimization framework based on heuristic and DQN is proposed. Finally, by taking into account the spatial and temporal distribution characteristics of satellite and facility resources, a multi-satellite scheduling instance dataset is generated. Experimental results validate the rationality and correctness of the DQN algorithm in solving the collaborative scheduling problem of multi-satellite data transmission.

For a multi-robot system, the accurate global map building based on a local map obtained by a single robot is an essential issue. The map building process is always divided into three stages: single-robot map acquisition, multi-robot map transmission, and multi-robot map merging. Based on the different stages of map building, this paper proposes a multi-stage optimization (MSO) method to improve the accuracy of the global map. In the map acquisition stage, we windowed the map based on the position of the robot to obtain the local map. Furthermore, we adopted the extended Kalman filter (EKF) to improve the positioning accuracy, thereby enhancing the accuracy of the map acquisition by the single robot. In the map transmission stage, considering the robustness of the multi-robot system in the real environment, we designed a dynamic self-organized communication topology (DSCT) based on the master and slave sketch to ensure the efficiency and accuracy of map transferring. In the map merging stage, multi-layer information filtering (MLIF) was investigated to increase the accuracy of the global map. We performed simulation experiments on the Gazebo platform and compared the result of the proposed method with that of classic map building methods. In addition, the practicability of this method has been verified on the Turtlebot3 burger robot. Experimental results proved that the MSO method improves the accuracy of the global map built by the multi-robot system.