Nowadays, the massive industrial data has effectively improved the performance of the data-driven deep learning Remaining Useful Life (RUL) prediction method. However, there are still problems of assigning fixed weights to features and only coarse-grained consideration at the sequence level. This paper proposes a Transformer-based end-to-end feature-level mask self-supervised learning method for RUL prediction. First, by proposing a fine-grained feature-level mask self-supervised learning method, the data at different time points under all features in a time window is sent to two parallel learning streams with and without random masks. The model can learn more fine-grained degradation information by comparing the information extracted by the two parallel streams. Instead of assigning fixed weights to different features, the abstract information extracted through the above process is invariable correlations between features, which has a good generalization to various situations under different working conditions. Then, the extracted information is encoded and decoded again using an asymmetric structure, and a fully connected network is used to build a mapping between the extracted information and the RUL. We conduct experiments on the public C-MAPSS datasets and show that the proposed method outperforms the other methods, and its advantages are more obvious in complex multi-working conditions.

An incubator is a kind of public or private organization that can provide multidimensional support to startup companies in exchange for profit or stake sharing. To date, huge numbers of incubators have been established all over the world, and many successful companies have been incubated by them. However, most of the existing incubators are more focused on incubating the target startup companies and pay relatively less attention to promoting the establishment of the supply chain networks and the ecosystems for the startup companies. In this regard, a new kind of socialized incubator model is established. Based on a kind of “Service-Product-Ecosystem” oriented configuration/operation mechanism, the socialized incubator can provide multidimension production services and value-added services during the entire incubation life-cycle of startup companies, transfer startup companies into ready-for-sold products, and construct multilayer vertical industrial ecosystems for the startup companies using socialized manufacturing resources. The socialized incubator model is verified through a case study on incubating a real startup company.

The clinical manifestations of OSA (obstructive sleep apnea) are night sleep snoring accompanied by apnea. fNIRS (functional near-infrared spectroscopy) can measure the relative concentration changes of oxyhemoglobin and deoxyhemoglobin, thereby measuring hemodynamics and oxygenation. Combined with the clinical manifestations of obstructive sleep apnea, fNIRS technology can be used in the monitoring of OSA. According to the principle of fNIRS, a miniaturized, wireless, and flexible device is designed in the article. The device can measure the change of blood oxygen concentration in the forehead of the brain, so that achieve the purpose of detecting OSA. To verify the effectiveness of the equipment, a breathing experiment was designed in the article to compare the equipment with the existing NIRX system. To our best knowledge, this is the first work that proposed a miniaturized flexible sleep apnea system based on monitoring fluctuations of cerebral hemodynamics. Meanwhile, the feasibility of the proposed system was verified. However, the stability of the equipment and related data processing algorithms need to be enhanced. The proposed system is expected to pave the way for the investigation of the involvement of the cerebral hemodynamics in the pathogenesis of OSA patients.

Deep reinforcement learning (DRL) has been a recent trend to solve autonomous driving decision-making (DM) problems. The goals of DM are to produce actions with ensuring safety and improving efficiency according to the information of perception and localization. Intuitively, the complexity and density of the driving scene will affect the DM while few works focus on it. This work is the first to integrate density characteristics in DRL. Specifically, we model three density indicators, 1) local density to describe the vehicle density relationship around the ego vehicle, 2) lane density to describe the denseness of vehicles on each lane, 3) global density to describe the denseness of current region of interest of DM. Among them, local density and lane density are considered as input features to make the agent aware of density properties to distinguish dense scenes while global density is used as a factor of the reward function to balance the safety and efficiency. The agent learns to make full use of density information to help the current DM process through learning. We train and test the models in an environment with different density scenarios. The experimental results show that considering density in DRL significantly helps the agent learn to make more secure and efficient decisions for autonomous driving.

Microservice architecture is applied by an increasing number of systems because of its benefits on delivery, scalability, and autonomy. It is essential but challenging to localize root-cause microservices promptly when a fault occurs. Traces are helpful for root-cause microservice localization, and thus many recent approaches utilize them. However, these approaches are less practical due to relying on supervision or other unrealistic assumptions. To overcome their limitations, we propose a more practical root-cause microservice localization approach named TraceRCA. The key insight of TraceRCA is that a microservice with more abnormal and less normal traces passing through it is more likely to be the root cause. Based on it, TraceRCA is composed of trace anomaly detection, suspicious microservice set mining and microservice ranking. We conducted experiments on hundreds of injected faults in a widely-used open-source microservice benchmark and a production system. The results show that TraceRCA is effective in various situations. The top-1 accuracy of TraceRCA outperforms the state-of-the-art unsupervised approaches by 44.8%. Besides, TraceRCA is applied in a large commercial bank, and it helps operators localize root causes for real-world faults accurately and efficiently. We also share some lessons learned from our real-world deployment.

This paper presents an energy-efficient and terrain-information-and-preceding-vehicle-information-incorporated energy management strategy for fully electric vehicles (FEVs) equipped with in-wheel motors. Saving driving energy with terrain preview and preceding vehicle movement prediction are crucial to prolong the driving distance for an FEV. Unlike conducting energy optimization under the assumption that the preceding vehicle movements are already known in most studies, the front vehicle movements are predicted during each control cycle based on the vehicle-to-vehicle communication, and the FEV vehicle velocity and motor torque distribution are optimized by a nonlinear model predictive controller to reduce energy consumption. The energy-saving objective is achieved by including, in the cost function, the motor energy consumption in each control cycle, while the safety objective is accomplished by keeping a suitable relative distance from the preceding vehicle. Since the nonlinear vehicle longitudinal model is applied, the gridding initial torque plane is utilized in each time step to search for the global minimum. Simulation results show that this method has a better energy-saving performance than the control method without using the preceding vehicle movement information, and the algorithm proposed here has a wide applicability under various driving conditions.

Adaptive cruise control (ACC) of hybrid electric vehicles (HEVs) has been traditionally developed without an efficient integration with active safety and energy management systems of hybrid power-trains, mainly for facilitating its implementation. This, however, leads to a compromise in the fuel economy of HEVs, since the predictive driving information provided by ACC is not exploited by the energy management system. In order to enhance the energy efficiency and control system integration, a novel ACC system for intelligent HEVs (i-HEV ACC) is developed in this study. The controller is proposed within the framework of nonlinear model predictive control, and a position-based nonlinear longitudinal intervehicle dynamics model is developed. A coordinated optimal control problem for both the tracking safety and the fuel consumption is formulated subject to the constraints on stable tracking. A multistep offline dynamic programming optimization and an online lookup table are used to implement the real-time control algorithm. Experiments are further conducted, which demonstrate that the proposed i-HEV ACC achieves enhanced performance and cooperation in traffic safety, fuel efficiency, and ride comfort.