一种用于机器人电池电量预测的Sarsa强化学习混合集成方法

建设数据高效互联的轨道交通车间已成为当前轨道交通装备行业转型发展的必然趋势。越来越多样化的移动运输机器人设备成为智能工厂数字化转型过程中的关键。准确预测机器人的电池电量可以指导控制中心提前采取科学合理的指令,确保物流运输链高效稳定运行。在本研究中,我们提出了一种基于状态-动作-奖励-状态-动作(Sarsa)强化学习算法的多学习器混合集成方法。首先,采用最大重叠离散小波变换(MODWT)对所测量的机器人原始电源电压数据进行预处理,可以显著降低时间序列数据的非平稳性和波动性。其次,利用门控循环单元(GRU)、深度置信网络(DBN)和长短期记忆(LSTM)对分解后得到的子序列进行预测建模。最后,使用Sarsa强化学习集成策略对上述三个基础预测器进行加权组合。所提出的Sarsa混合集成模型的性能在三个真实移动机器人功率数据集上得到验证。实验结果表明,运输机器人电池动力混合预测模型在鲁棒性、准确性和适应性方面具有竞争力。

Correcting of the unexpected localization measurement for indoor automatic mobile robot transportation based on a neural network

The increasing use of mobile robots in laboratory settings has led to a higher degree of laboratory automation.However,when mobile robots move in laboratory environments,mechanical errors,environmental disturbances and signal interruptions are inevitable.This can compromise the accuracy of the robot’s localization,which is crucial for the safety of staff,robots and the laboratory.A novel time-series predicting model based on the data processing method is proposed to handle the unexpected localization measurement of mobile robots in laboratory environments.The proposed model serves as an auxiliary localization system that can accurately correct unexpected localization errors by relying solely on the historical data of mobile robots.The experimental results demonstrate the effectiveness of this proposed method.

An enhanced hybrid ensemble deep learning approach for forecasting daily PM_(2.5)

PM_(2.5) forecasting technology can provide a scientific and effective way to assist environmental governance and protect public health.To forecast PM_(2.5),an enhanced hybrid ensemble deep learning model is proposed in this research.The whole framework of the proposed model can be generalized as follows:the original PM_(2.5) series is decomposed into 8 sub-series with different frequency characteristics by variational mode decomposition(VMD);the long short-term memory(LSTM)network,echo state network(ESN),and temporal convolutional network(TCN)are applied for parallel forecasting for 8 different frequency PM_(2.5) sub-series;the gradient boosting decision tree(GBDT)is applied to assemble and reconstruct the forecasting results of LSTM,ESN and TCN.By comparing the forecasting data of the models over 3 PM_(2.5) series collected from Shenyang,Changsha and Shenzhen,the conclusions can be drawn that GBDT is a more effective method to integrate the forecasting result than traditional heuristic algorithms;MAE values of the proposed model on 3 PM_(2.5) series are 1.587,1.718 and 1.327μg/m3,respectively and the proposed model achieves more accurate results for all experiments than sixteen alternative forecasting models which contain three state-of-the-art models.

A hybrid decomposition-boosting model for short-term multi-step solar radiation forecasting with NARX neural network

Due to global energy depletion,solar energy technology has been widely used in the world.The output power of the solar energy systems is affected by solar radiation.Accurate short-term forecasting of solar radiation can ensure the safety of photovoltaic grids and improve the utilization efficiency of the solar energy systems.In the study,a new decomposition-boosting model using artificial intelligence is proposed to realize the solar radiation multi-step prediction.The proposed model includes four parts:signal decomposition(EWT),neural network(NARX),Adaboost and ARIMA.Three real solar radiation datasets from Changde,China were used to validate the efficiency of the proposed model.To verify the robustness of the multi-step prediction model,this experiment compared nine models and made 1,3,and 5 steps ahead predictions for the time series.It is verified that the proposed model has the best performance among all models.

BLE beacons for sample position estimation in a life science automation laboratory

Estimation of the sample position is essential for working process monitoring and management in the life science automation laboratory.Bluetooth low-energy(BLE)beacons have the advantages of low price,small size and low energy consumption,which make them a promising solution for sample position estimation in the automated laboratory.Several fingerprinting models have been proposed to achieve indoor localization with the received signal strength(RSS)data.However,most of the research depends on intensive beacon installation.Proximity estimation,which depends entirely on one beacon,is more suitable for sample position estimation in large automated laboratories.The complexity of the life science automation laboratory environment brings challenges to the traditional path loss model(PLM),which is a widely used radio wave propagation model-based proximity estimation method.In this paper,BLE sensing devices for sample position estimation are proposed.The BLE beacon-based proximity estimation is discussed in the framework of machine learning,in which the support vector regression(SVR)is utilized to model the nonlinear relationship between the RSS data and distance,and the Kalman filter is utilized to decrease the RSS data deviation.The experimental results over different environments indicate that the SVR outperforms the PLM significantly,and provides 1 m absolute errors for more than 95%of the testing samples.The Kalman filter brings benefits to stable distance predictions.Apart from proximity-based sample position estimation,the proposed framework turned out to be effective in position estimation between parallel workbenches and position estimation on an automated workstation.

BLE beacon-based floor detection for mobile robots in a multi-floor automation laboratory

As an important task of multi-floor localization,floor detection has elicited great attention.Wireless infrastructures like Wi-Fi and Bluetooth Low Energy(BLE)play important roles in floor detection.However,most floor detection research tends to focus on data modelling but pays little attention to the data collection system,which is the basis of wireless infrastructure-based floor detection.In fact,the floor detection task can be greatly simplified with proper data collection system design.In this paper,a floor detection solution is developed in a multi-floor life science automation lab.A data collection system consisting of BLE beacons,a receiver node and an Internet of Things(IoT)cloud is provided.The features of the BLE beacon under different settings are evaluated in detail.A mean filter is designed to deal with the fluctuation of the received signal strength indicator data.A simple floor detection method without a training process was implemented and evaluated in more than 100 floor detection tests.The time delay and floor detection accuracy under different settings are discussed.Finally,floor detection is evaluated on the H20 multi-floor transportation robot.Two sensor nodes are installed on the robot at different heights.The floor detection performance with different installation heights is discussed.The experimental results indicate that the proposed floor detection method provides floor detection accuracy of 0.9877 to 1 with a time delay of 5s.

Research on Modeling and Fuzzy Control of Magneto-Rheological Intelligent Buffer System for Impact Load

So far the magneto-rheological(MR) effect mechanism of MR damper has not been known completely, especially in the impact load,and the problem becomes more complicated and difficult for analyzing.A set of characteristic tests and parameters' identification are made to the MR damper by the experimental platform. The dynamical model of the damper is constructed based on the Bingham plastic model,and the buffer control strategy of aircraft undercarriage based on MR technology is established.Finally,the fuzzy control algorithm is applied to the process of automatic control for landing buffer of aircraft undercarriage.The simulation results show that the proposed MR damper pulley buffer can effectively recognize the impact energy.The research has a better application in the engineering.

Deep Learning for Visual SLAM in Transportation Robotics:A review

Visual SLAM(Simultaneously Localization and Mapping)is a solution to achieve localization and mapping of robots simultaneously.Significant achievements have been made during the past decades,geography-based methods are becoming more and more successful in dealing with static environments.However,they still cannot handle a challenging environment.With the great achievements of deep learning methods in the field of computer vision,there is a trend of applying deep learning methods to visual SLAM.In this paper,the latest research progress of deep learning applied to the field of visual SLAM is reviewed.The outstanding research results of deep learning visual odometry and deep learning loop closure detect are summarized.Finally,future development directions of visual SLAM based on deep learning is prospected.

Modeling and Control of the Pulley Buffer System of Arresting Cable for Shipboard Aircraft Based on Magneto-Rheological Fluid

Taking the MK7-3 of USA hydraulic buffer arresting device as the research subject,the dynamical model for the shipboard aircraft arresting system is established,and the magneto-rheological(MR) damper is applied to pulley shock absorbers for shipboard aircraft block system.Due to the effect of the MR damper has not been known completely and so far MR damper model has not been defined,we use a set of characteristic test of the MR damper,through the process of parameters identification,to establish the dynamical model for the MR damper based on the Bingham plastic model.Then,the fuzzy control rules are designed,the buffer control for the pulley buffer of shipboard aircrafts is completed in touchdown moment based on MR technology. Compared with blocking device of hydraulic pulley buffer in the same condition,the simulations results show that the proposed MR pulley buffer can effectively recognize the impact energy for shipboard block system and reduce the pull peak of arresting cable.It improves significantly safety during landing of the air vehicles and lowers the risk of accidents.

Railway switch fault diagnosis based on Multi-heads Channel Self Attention,Residual Connection and Deep CNN

A novel switch diagnosis method based on self-attention and residual deep convolutional neural networks(CNNs)is proposed.Because of the imbalanced dataset,the K-means synthetic minority oversampling technique(SMOTE)is applied to balancing the dataset at first.Then,the deep CNN is utilized to extract local features from long power curves,and the residual connection is performed to handle the performance degeneration.In the end,the multi-heads channel self attention focuses on those important local features.The ablation and comparison experiments are applied to verifying the effectiveness of the proposed methods.With the residual connection and multi-heads channel self attention,the proposed method has achieved an impressive accuracy of 99.83%.The t-SNE based visualizations for features of the middle layers enhance the trustworthiness.

A hybrid ensemble deep reinforcement learning model for locomotive axle temperature using the deterministic and probabilistic strategy

This paper proposes a hybrid deep reinforcement learning framework for locomotive axle temperature by combining the wavelet packet decomposition(WPD),long short-term memory(LSTM),gated recurrent unit(GRU)reinforcement learning and generalized autoregressive conditional heteroskedasticity(GARCH)algorithms.The WPD is utilized to decompose the raw nonlinear series into subseries.Then the deep learning predictors LSTM and GRU are established to predict the future axle temperatures in each subseries.The Q-learning could generate optimal ensembleweights to integrate the predictors to finish the deterministic forecasting and GARCH is used to conduct the deterministic forecasting based on the deterministic forecasting residual.These parts of the hybrid ensemble structure contributed to optimal modelling accuracy and provided effective support in the real-time monitoring and fault diagnosis of transportation.

Multi-floor laboratory transportation technologies based on intelligent mobile robots

Transportation technologies for mobile robots include indoor navigation,intelligent collision avoidance and target manipulation.This paper discusses the research process and development of these interrelated technologies.An efficient multi-floor laboratory transportation system for mobile robots developed by the group at the Center for Life Science Automation(CELISCA)is then introduced.This system is integrated with the multi-floor navigation and intelligent collision avoidance systems,as well as a labware manipulation system.A multi-floor navigation technology is proposed,comprising sub-systems for mapping and localization,path planning,door control and elevator operation.Based on human-robot interaction technology,a collision avoidance system is proposed that improves the navigation of the robots and ensures the safety of the transportation process.Grasping and placing operation technologies using the dual arms of the robots are investigated and integrated into the multi-floor transportation system.The proposed transportation system is installed on the H20 mobile robots and tested at the CELISCA laboratory.The results show that the proposed system can ensure the mobile robots are successful when performing multi-floor laboratory transportation tasks.