Heart rate (HR) is an important indicator reflecting the overall physical and mental health of the human body, playing a crucial role in diagnosing cardiovascular and neurological diseases. Recent research has revealed that variations in the light absorption of human skin captured through facial video over the cardiac cycle, due to changes in blood volume, can be utilized for non-contact HR estimation. However, most existing methods rely on single-modal video sources (such as RGB or NIR), which often yield suboptimal results due to noise and the limitations of a single information source. To overcome these challenges, this paper proposes a multimodal information fusion architecture named the spatiotemporal sensitive network (SS-Net) for non-contact heart rate estimation. Firstly, spatiotemporal feature maps are utilized to extract physiological signals from RGB and NIR videos effectively. Next, a spatiotemporal sensitive (SS) module is introduced to extract useful physiological signal information from both RGB and NIR spatiotemporal maps. Finally, a multi-level spatiotemporal context fusion (MLSC) module is designed to fuse and complement information between the visible light and infrared modalities. Then, different levels of fused features are refined in task-specific branches to predict both remote photoplethysmography (rPPG) signals and heart rate (HR) signals. Experiments conducted on three datasets demonstrate that the proposed SS-Net achieves superior performance compared to existing methods.

More and more robot tasks extend to unknown environments. Extraction of the topological map from the robot exploration results of an unknown environment is of great importance for the completion of robot path-planning tasks. In this research, a robot explores the environment and builds a grid map in the simulation experiments; the building of topological map based on the features of the grid map is studied. A new method is proposed that the grid environment be classified to 8 basic types of concave and protruding corners and the 8 basic types be defined by mathematical equations. According to the definition a strategy is designed to divide the grid map into s and a topological map of the environment is built

The fault diagnosis and inspection policy were discussed for a system which has two operating states (normal, abnormal) and a failure state. The system is inspected periodically. The purpose of inspection is to identify the state of system and adopt appropriate repair strategy by measuring the value of diagnosis parameter. Because the system is deteriorated gradually and a replacement is expensive, several imperfect repairs are allowed until a replacement is needed. The objective of this paper is to determine the optimal critical value of diagnosis parameter and the optimal inspection cycle such that the expected profit per unit in steady states is maximized. Reliability indexes of the system are also derived. Probability analysis and vector Markov process are used in this paper. Researches on the numerical example show the feasibility of the policy.