In the context of intelligent coal mining faces, the strapdown inertial navigation system (SINS) serves as a pivotal technology for realizing the autonomous positioning of shearers. However, due to the complex underground environment, the positioning precision and stability of SINS have yet to meet the requirements for the normalized operation of intelligent mining faces. This limitation severely impedes the widespread application of SINS in coal mines. To address these challenges and enhance the performance of shearer SINS, this paper proposes a positioning accuracy improvement method based on the SINS and optical fiber sensing system (OFSS). By leveraging the spatial consistency between the shearer trajectory and the shape curve of the scraper conveyor, a SINS/OFSS integrated positioning model tailored to the shearer’s motion characteristics is established. Furthermore, a position mapping mechanism is introduced to achieve spatiotemporal consistency between the real-time position of the shearer and the discrete points on the scraper conveyor’s shape curve. Subsequently, a shearer trajectory time-series prediction model is constructed using historical states to predict future trajectories. This model serves to evaluate and perform a secondary correction on the positioning results, thereby further strengthening the error correction capability. Finally, simulation experiments and platform tests are conducted to validate the effectiveness of the proposed method. The results indicate that under both normal cutting and oblique cutting conditions, the positioning precision and stability of the proposed method significantly outperform comparative methods. The combination of the SINS/OFSS integrated positioning model and the TCN-BiLSTM-Attention (TBA) prediction model effectively mitigates SINS positioning errors, realizing a substantial improvement in shearer positioning accuracy.

Human activity recognition (HAR) is a task designed to identify and classify physical activities or behaviors in people’s daily lives. This field relies on data collected from various sensors. Due to differences in environment, user posture and habits, as well as equipment placement, the data exhibits severe heterogeneity. The channel information from different sensors exhibits more complex and uncertain characteristics in terms of shape, noise level, and data quality. These properties can lead to poor generalization of deep learning models, posing challenges for the effectiveness of deep learning algorithms and the widespread use of specific embedded devices in this field. Therefore, this article proposes an adaptive channel signal scaling method to calibrate channel characteristics from the perspective of sensor channel information. Additionally, we propose a stable feature completion strategy to enrich the data feature information using different fusion strategies at the data level. Extensive experiments were conducted on three publicly available HAR datasets, and the experimental results demonstrate that our proposed method significantly improves the performance of state-of-the-art deep learning methods.

Federated Learning (FL) is an emerging privacy-aware machine learning technique that applies successfully to the collaborative learning of global models for Human Activity Recognition (HAR). As of now, the applications of FL for HAR assume that the data associated with diverse individuals follow the same distribution. However, this assumption is impractical in real-world scenarios where the same activity is frequently performed differently by different individuals. To tackle this issue, we propose FedMAT, a Federated Multi-task ATtention framework for HAR, which extracts and fuses shared as well as individual-specific multi-modal sensor data features. Specifically, we treat the HAR problem associated with each individual as a different task and train a federated multi-task model, composed of a shared feature representation network in a central server plus multiple individual-specific networks with attention modules stored in decentralized nodes. In this architecture, the attention module operates as a mask that allows to learn individual-specific features from the global model, whilst simultaneously allowing for features to be shared among different individuals. We conduct extensive experiments based on publicly available HAR datasets, which are collected in both controlled environments and real-world scenarios. Numeric results verify that our proposed FedMAT significantly outperforms baselines not only in generalizing to existing individuals but also in adapting to new individuals.