Daoqu Geng

Efficient memory management is essential for the stability and long-term performance of mobile robots in Simultaneous Localization and Mapping (SLAM). However, existing methods often struggle to control redundancy in keyframes and map points, leading to reduced efficiency, increased latency, and potential system failure due to resource constraints. Achieving high accuracy in both mapping and trajectory estimation while maintaining a compact state representation remains a key challenge for scalable and efficient SLAM systems. To address this issue, this paper proposes an efficient long-term visual SLAM method based on sparse prior embedding and nonlinear score-guided sparsification for memory-constrained environments. The approach embeds keyframe information into sparse prior factors, avoiding global coupling while preserving system sparsity and consistency. Additionally, a nonlinear scoring function combining parallax and descriptor uniqueness is introduced to guide map point sparsification within the sliding window. This strategy enables efficient state graph management, achieving compact global map representations and effective observation constraints. The proposed method has been implemented in a complete visual SLAM system and evaluated through long-term real-world mapping experiments on an embedded robotic platform. Experimental results demonstrate that the approach significantly reduces memory consumption while maintaining trajectory and mapping accuracy. Furthermore, the method ensures real-time execution and deployment potential, indicating its suitability for large-scale SLAM tasks in resource-constrained and long-duration operational scenarios.

The recent global trend is the convergence of information and communications technology (ICT). By applying ICT in various fields such as the humanities, new types of products and services are created, and new values that help people's lives can be created. AI can be selected as a representative technology in such convergence ICT. However, applying AI technology to actual production requires ensuring data security. Federated learning (FL) can achieve secure sharing of data, where all parties participate in model training locally and upload it to the server for aggregation. The data never leaves the parties involved, thus solving the problems of data privacy and data silos. However, FL faces issues such as high communication cost, imbalanced performance distribution among participants, and low privacy protection. To achieve a balance between model accuracy, communication cost, fairness, and privacy, this paper proposes a multi‐objective optimization‐based FL algorithm (M‐FedAvg). The multi‐objective optimization problem of maximising the accuracy of the global model, minimising the communication cost, minimising the variance of the accuracy, and minimising the privacy budget is solved by NSGA‐III. The experimental results show that the algorithm proposed can effectively reduce the communication cost of FL and achieve privacy protection for participants without affecting the accuracy of the global model.

Naïve Bayes classifiers have made a big splash in the field of data mining and supervised learning. However, the assumption of "conditional independence of attributes" is difficult to realize in reality, which affects the final classification results. To address this problem, this paper introduces a dynamic weighting mechanism to assign corresponding weights to each attribute according to its influence on the classification result, and constructs a weight-based naïve Bayes classifier. The classifier takes the classification error rate as the objective function, uses the Harmony Search Algorithm to calculate the global optimal weights, and assigns the weights to each attribute after normalization. To improve the optimal-seeking ability of the Harmony Search Algorithm, this paper optimizes its new solution generation method and reduces the probability of falling into a local optimum by using an optimization scheme that accepts worse solutions with a certain probability. Experimental results demonstrate that the algorithm proposed in this paper has some improvement in classification indicators such as correct rate and F1-Score compared with the classical naïve Bayes classifier and other optimized naïve Bayes classifiers.

Big data-based acquisition and storage system (ASS) plays an important role in the design of industrial data platform. Many big data frameworks have been integrated compression and serialization method. These methods cannot meet the needs of industrial production information management for requiring time-consuming and mass storage. Based on the existing big data frameworks, we propose an enhanced industrial big data platform in order to reduce the data processing time while requiring fewer data storage space. Specifically, this paper focuses on evaluating the impact of multiple compression and serialization methods on big data platform performance and tries to choose optimal compression and serialization methods for the industrial data platform. Compared to the methods integrated in Hadoop and Spark, the experimental results showed the data compression time of the platform has been reduced by 73.9% with a less than 96% the size of data compressed, furthermore, the data serialization time has been reduced by 80.8%. With the increasing amount of data, it takes less time to compare with benchmark methods.

We present a novel and practical micro weather station, which can sense temperature, relative humidity, pressure and anemometer, and is portable in small size and possesses high precisions. The micro weather station comprises multi-sensor chip, anemometer, measurement system, display system and power management system. Based on MEMS technology, multi-sensor chip integrated temperature, relative humidity and pressure is developed and manufactured. A drag force wind sensor using the torque of cantilever to measure the velocity of wind is developed. The wind direction can be measured by perpendicularly encapsulating the two wind sensor. Compared with those processes used in other types of micro weather station, the processes we used were very simple and compatible. All the results exhibit outstanding performances of our micro weather station.