This study focuses on the motion planning and risk evaluation of unprotected left turns at occluded intersections for autonomous vehicles. In this paper, we present an interactive motion planning controller that combines Cross-Observability Optimistic-Pessimistic Safe Reinforcement Learning (COOP-SRL) and Nonlinear Model Predictive Control (NMPC), with consideration of the uncertain potential risk of occluded zone, the trade-off between safety and efficiency, and the dynamic interaction between vehicles. The proposed COOP-SRL algorithm integrates fully and partially observable policies through cross-observability soft imitation learning to leverage the expert guidance and improve learning efficiency. Moreover, the optimistic exploration policy and pessimism safe constraint are adopted to provide an adaptive safe strategy without hindering the exploration during learning process. Finally, the evaluations of the proposed controller were conducted in occluded intersection scenarios with various traffic density level, which indicate that the proposed method outperforms both the optimization-based and learning-based baselines in qualitative and quantitative indexes.
{"title":"Cross-Observability Optimistic-Pessimistic Safe Reinforcement Learning for Interactive Motion Planning With Visual Occlusion","authors":"Xiaohui Hou;Minggang Gan;Wei Wu;Yuan Ji;Shiyue Zhao;Jie Chen","doi":"10.1109/TITS.2024.3443397","DOIUrl":"https://doi.org/10.1109/TITS.2024.3443397","url":null,"abstract":"This study focuses on the motion planning and risk evaluation of unprotected left turns at occluded intersections for autonomous vehicles. In this paper, we present an interactive motion planning controller that combines Cross-Observability Optimistic-Pessimistic Safe Reinforcement Learning (COOP-SRL) and Nonlinear Model Predictive Control (NMPC), with consideration of the uncertain potential risk of occluded zone, the trade-off between safety and efficiency, and the dynamic interaction between vehicles. The proposed COOP-SRL algorithm integrates fully and partially observable policies through cross-observability soft imitation learning to leverage the expert guidance and improve learning efficiency. Moreover, the optimistic exploration policy and pessimism safe constraint are adopted to provide an adaptive safe strategy without hindering the exploration during learning process. Finally, the evaluations of the proposed controller were conducted in occluded intersection scenarios with various traffic density level, which indicate that the proposed method outperforms both the optimization-based and learning-based baselines in qualitative and quantitative indexes.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"17602-17613"},"PeriodicalIF":7.9,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142579164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-23DOI: 10.1109/TITS.2024.3424808
Junfeng Zhang;Cheng Xie;Hongming Cai;Weiming Shen;Rui Yang
Real-Time Traffic Flow Prediction (RT-TFP) is one of the critical technologies for implementing the Intelligent Transportation System (ITS), enabling rapid and accurate prediction of real-time traffic flow at intersections. RT-TFP typically needs to be deployed on-site edge devices for real-time traffic flow calculation that requires low inference latency and minimal computational resources. However, the existing Traffic Flow Prediction (TFP) models are generally based on spatiotemporal graph neural networks (STGNNs), which are complex and require high computational resources and relatively high inference times that can hardly be deployed on edge devices. To this end, this work proposes a simple RT-TFP model, SpatioTemporal-MultiLayer Perceptron (ST-MLP), which requires low computational resources and inference times. The base idea of this work is to establish a spatio-temporal MLP model to replace the STGNN model for conducting the TFP, which is much faster and simpler. Specifically, first, a TempEncoder is proposed to encode the temporal information into the MLP features. Then, a Spatiotemporal Mixer is proposed to mix spatial information into the temporal-enriched MLP features. After, MLP features are distilled from a complex STGNN model to obtain a simple MLP that inherits complete Spatial-Temporal information of the traffic graph. The experimental results on four real-world datasets show the proposed model achieves competitive prediction accuracy with STGNN models in much fewer computational resources and lower prediction time costs. It is worth noting that, the proposed method is faster than the compared STGNNs by an average of 21.62 times (~10.81s $rightsquigarrow ~sim 0.50$ s). Interestingly, the proposed ST-MLP even has a −3.23% error rate decreasing on average compared to the corresponding STGNN model. Moreover, the error rate of the proposed ST-MLP decreases over pure MLPs by −3.92% $sim -42.62$ %. The source code is available at: https://github.com/zhangjunfeng1234/ST-MLP
{"title":"Knowledge Distillation-Based Spatio-Temporal MLP Model for Real-Time Traffic Flow Prediction","authors":"Junfeng Zhang;Cheng Xie;Hongming Cai;Weiming Shen;Rui Yang","doi":"10.1109/TITS.2024.3424808","DOIUrl":"https://doi.org/10.1109/TITS.2024.3424808","url":null,"abstract":"Real-Time Traffic Flow Prediction (RT-TFP) is one of the critical technologies for implementing the Intelligent Transportation System (ITS), enabling rapid and accurate prediction of real-time traffic flow at intersections. RT-TFP typically needs to be deployed on-site edge devices for real-time traffic flow calculation that requires low inference latency and minimal computational resources. However, the existing Traffic Flow Prediction (TFP) models are generally based on spatiotemporal graph neural networks (STGNNs), which are complex and require high computational resources and relatively high inference times that can hardly be deployed on edge devices. To this end, this work proposes a simple RT-TFP model, SpatioTemporal-MultiLayer Perceptron (ST-MLP), which requires low computational resources and inference times. The base idea of this work is to establish a spatio-temporal MLP model to replace the STGNN model for conducting the TFP, which is much faster and simpler. Specifically, first, a TempEncoder is proposed to encode the temporal information into the MLP features. Then, a Spatiotemporal Mixer is proposed to mix spatial information into the temporal-enriched MLP features. After, MLP features are distilled from a complex STGNN model to obtain a simple MLP that inherits complete Spatial-Temporal information of the traffic graph. The experimental results on four real-world datasets show the proposed model achieves competitive prediction accuracy with STGNN models in much fewer computational resources and lower prediction time costs. It is worth noting that, the proposed method is faster than the compared STGNNs by an average of 21.62 times (~10.81s \u0000<inline-formula> <tex-math>$rightsquigarrow ~sim 0.50$ </tex-math></inline-formula>\u0000s). Interestingly, the proposed ST-MLP even has a −3.23% error rate decreasing on average compared to the corresponding STGNN model. Moreover, the error rate of the proposed ST-MLP decreases over pure MLPs by −3.92% \u0000<inline-formula> <tex-math>$sim -42.62$ </tex-math></inline-formula>\u0000%. The source code is available at: \u0000<uri>https://github.com/zhangjunfeng1234/ST-MLP</uri>","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"18122-18135"},"PeriodicalIF":7.9,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142587524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/TITS.2024.3439618
Kai Ma;Zhenyu Liu;Yuan Shen
Multi-agent sensing for transportation systems is receiving widespread attention due to its dynamic flexibility and collaborative capabilities, where the target sensing error is limited by the spatio-temporal error caused by agent localization and formation steps. This paper considers the sensing problem of non-cooperative targets (UAVs or vehicles) by cooperative asynchronous agents (UAVs). This paper develops a framework where the formation of agents and the allocation of resources are jointly optimized. In particular, we reveal the error coupling of measurement and motion noises on target sensing accuracy by Fisher information analysis. Then we propose bandwidth allocation and agent activation strategies in the localization step, which simultaneously improve the position accuracy of agents and the quality of sensing signals. In the formation step, we design motion planning algorithms to increase sensing information about targets. Simulation results demonstrate the significant performance improvements achieved by our proposed algorithms that minimize the effects of localization and control errors on target sensing.
{"title":"A Multi-Agent Sensing Framework via Joint Motion Planning and Resource Optimization","authors":"Kai Ma;Zhenyu Liu;Yuan Shen","doi":"10.1109/TITS.2024.3439618","DOIUrl":"10.1109/TITS.2024.3439618","url":null,"abstract":"Multi-agent sensing for transportation systems is receiving widespread attention due to its dynamic flexibility and collaborative capabilities, where the target sensing error is limited by the spatio-temporal error caused by agent localization and formation steps. This paper considers the sensing problem of non-cooperative targets (UAVs or vehicles) by cooperative asynchronous agents (UAVs). This paper develops a framework where the formation of agents and the allocation of resources are jointly optimized. In particular, we reveal the error coupling of measurement and motion noises on target sensing accuracy by Fisher information analysis. Then we propose bandwidth allocation and agent activation strategies in the localization step, which simultaneously improve the position accuracy of agents and the quality of sensing signals. In the formation step, we design motion planning algorithms to increase sensing information about targets. Simulation results demonstrate the significant performance improvements achieved by our proposed algorithms that minimize the effects of localization and control errors on target sensing.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"18925-18938"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/TITS.2024.3453495
Lijuan Wan;Chunhui Yu;Hong K. Lo
Stochastic traffic demands and turning ratios are critical factors in coordinated signal control. However, existing studies ignore the spatial-temporal dependencies of traffic flows between adjacent intersections and signal cycles. Turning ratios are usually assumed to be deterministic. This study develops a two-stage stochastic programming model for two-way coordinated adaptive signal control under stochastic traffic demands and turning ratios. A hierarchical multi-objective function is developed for overflow management and operational efficiency under both over- and under-saturated traffic. The primary and secondary objective functions minimize residual queue lengths and average vehicle delays, respectively, which are formulated considering spatial-temporal dependencies for the coordinated traffic flow. In stage one, a base coordinated signal timing plan is optimized to maximize the expected performance under stochastic scenarios. In stage two, adaptive cycle lengths and green times are determined by setting the tolerance factor for the base green times to maintain the stable traffic flow. The concept of Phase Clearance Reliability (PCR) is extended to decouple the interaction between the two stages. The deterministic equivalent problem of the proposed model in one signal cycle is modified to optimize the base signal timing plan for serving the stochastic exogenous and endogenous traffic demands up to certain PCR values. A PCR-based gradient algorithm is designed for solutions. The experimental results demonstrate that the proposed model can significantly improve traffic operation compared to six benchmarks.
{"title":"Traffic Signal Coordination Under Stochastic Demands and Turning Ratios Considering Spatial-Temporal Dependencies","authors":"Lijuan Wan;Chunhui Yu;Hong K. Lo","doi":"10.1109/TITS.2024.3453495","DOIUrl":"10.1109/TITS.2024.3453495","url":null,"abstract":"Stochastic traffic demands and turning ratios are critical factors in coordinated signal control. However, existing studies ignore the spatial-temporal dependencies of traffic flows between adjacent intersections and signal cycles. Turning ratios are usually assumed to be deterministic. This study develops a two-stage stochastic programming model for two-way coordinated adaptive signal control under stochastic traffic demands and turning ratios. A hierarchical multi-objective function is developed for overflow management and operational efficiency under both over- and under-saturated traffic. The primary and secondary objective functions minimize residual queue lengths and average vehicle delays, respectively, which are formulated considering spatial-temporal dependencies for the coordinated traffic flow. In stage one, a base coordinated signal timing plan is optimized to maximize the expected performance under stochastic scenarios. In stage two, adaptive cycle lengths and green times are determined by setting the tolerance factor for the base green times to maintain the stable traffic flow. The concept of Phase Clearance Reliability (PCR) is extended to decouple the interaction between the two stages. The deterministic equivalent problem of the proposed model in one signal cycle is modified to optimize the base signal timing plan for serving the stochastic exogenous and endogenous traffic demands up to certain PCR values. A PCR-based gradient algorithm is designed for solutions. The experimental results demonstrate that the proposed model can significantly improve traffic operation compared to six benchmarks.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"18236-18251"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/tits.2024.3451509
Nan Xu, Min Hu, Lingge Jin, Haitao Ding, Yanjun Huang
{"title":"Combined-Slip Trajectory Tracking and Yaw Stability Control for 4WID Autonomous Vehicles Based on Effective Cornering Stiffness","authors":"Nan Xu, Min Hu, Lingge Jin, Haitao Ding, Yanjun Huang","doi":"10.1109/tits.2024.3451509","DOIUrl":"https://doi.org/10.1109/tits.2024.3451509","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"50 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/TITS.2024.3444853
James Klupacs;Amirali K. Gostar;Alireza Bab-Hadiashar;Reza Hoseinnezhad
This paper introduces an approach to distributed multi-object tracking for connected vehicles, aiming to overcome the inherent challenges of label inconsistency and double counting prevalent in distributed information fusion methods, particularly in the context of situational awareness for connected vehicles. Our proposed method expands the label space by incorporating sensor identity into the object’s label. Furthermore, we present an intuitive merging algorithm designed to effectively eliminate instances of double counting. The approach is formulated, and an algorithm is developed for implementation within a labeled multi-Bernoulli filter, executed locally on each node of a distributed network responsible for information fusion. To assess the efficacy of our solution, we evaluate its performance in a highly demanding scenario specifically designed for intelligent transport systems and compare its performance against alternative approaches.
{"title":"Enhancing Multi-Object Tracking Through Distributed Information Fusion in Connected Vehicle Networks","authors":"James Klupacs;Amirali K. Gostar;Alireza Bab-Hadiashar;Reza Hoseinnezhad","doi":"10.1109/TITS.2024.3444853","DOIUrl":"10.1109/TITS.2024.3444853","url":null,"abstract":"This paper introduces an approach to distributed multi-object tracking for connected vehicles, aiming to overcome the inherent challenges of label inconsistency and double counting prevalent in distributed information fusion methods, particularly in the context of situational awareness for connected vehicles. Our proposed method expands the label space by incorporating sensor identity into the object’s label. Furthermore, we present an intuitive merging algorithm designed to effectively eliminate instances of double counting. The approach is formulated, and an algorithm is developed for implementation within a labeled multi-Bernoulli filter, executed locally on each node of a distributed network responsible for information fusion. To assess the efficacy of our solution, we evaluate its performance in a highly demanding scenario specifically designed for intelligent transport systems and compare its performance against alternative approaches.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"15897-15908"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/tits.2024.3453340
Yuqing Wang, Junwei Zhang, Zhuo Ma, Ning Lu, Teng Li, Jianfeng Ma
{"title":"Location-Aware and Privacy-Preserving Data Cleaning for Intelligent Transportation","authors":"Yuqing Wang, Junwei Zhang, Zhuo Ma, Ning Lu, Teng Li, Jianfeng Ma","doi":"10.1109/tits.2024.3453340","DOIUrl":"https://doi.org/10.1109/tits.2024.3453340","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"6 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264169","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nowadays, Connected Automated Vehicles (CAVs) have emerged as powerful infrastructures for the next-generation Intelligent Transportation System (ITS) as the rapid technological advancements of communication networks and vehicular intelligence. While prospective platoon-based techniques in CAVs, the heterogeneous traffic condition poses a challenge for platoon control in the self-organized traffic bottleneck, thus making an urgent need for a practical sustainable transportation architecture. To address this problem, we propose a software defined architecture that leverages multi-agent techniques to mobile-edge computing networks for multi-vehicle adaptive platoon, which is called SD-M3ASP. The architecture supports centralized and decentralized management of vehicular edge communication resources between mobile vehicles and edge devices, and underpins sustainable vehicular platooning capabilities. Then, we propose cluster-based kinematic models by grouping vehicles into multi-vehicle clusters (MVCs) to facilitate efficient platoon control with collision avoidance. Furthermore, we propose three-stage platoon control algorithms to adaptively balance the size of MVCs and form stable platoons in heterogeneous traffic flows. The intra-platoon and inter-platoon convergence are analyzed by using the Routh stability criterion and Lyapunov technique. A CAV simulation software is developed for demonstration purposes which is available online at https://qgailab.com/cav-sim . Extensive numerical simulation results have shown the superiority of the proposed method, which can greatly eliminate the self-organized congestion caused by heterogeneous traffic flow.
{"title":"Sustainable Distributed Adaptive Platoon in Multi-Agent Mobile-Edge Computing Networks for Lane Reduction Scenario","authors":"Guangqiang Xie;Biwei Zhong;Haoran Xu;Yang Li;Xianbiao Hu;Zhihao Jiang;Yonghong Tian","doi":"10.1109/TITS.2024.3449916","DOIUrl":"10.1109/TITS.2024.3449916","url":null,"abstract":"Nowadays, Connected Automated Vehicles (CAVs) have emerged as powerful infrastructures for the next-generation Intelligent Transportation System (ITS) as the rapid technological advancements of communication networks and vehicular intelligence. While prospective platoon-based techniques in CAVs, the heterogeneous traffic condition poses a challenge for platoon control in the self-organized traffic bottleneck, thus making an urgent need for a practical sustainable transportation architecture. To address this problem, we propose a software defined architecture that leverages multi-agent techniques to mobile-edge computing networks for multi-vehicle adaptive platoon, which is called SD-M3ASP. The architecture supports centralized and decentralized management of vehicular edge communication resources between mobile vehicles and edge devices, and underpins sustainable vehicular platooning capabilities. Then, we propose cluster-based kinematic models by grouping vehicles into multi-vehicle clusters (MVCs) to facilitate efficient platoon control with collision avoidance. Furthermore, we propose three-stage platoon control algorithms to adaptively balance the size of MVCs and form stable platoons in heterogeneous traffic flows. The intra-platoon and inter-platoon convergence are analyzed by using the Routh stability criterion and Lyapunov technique. A CAV simulation software is developed for demonstration purposes which is available online at \u0000<uri>https://qgailab.com/cav-sim</uri>\u0000. Extensive numerical simulation results have shown the superiority of the proposed method, which can greatly eliminate the self-organized congestion caused by heterogeneous traffic flow.","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"25 11","pages":"15673-15686"},"PeriodicalIF":7.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1109/tits.2024.3451714
Weibin Zhang, Huazhu Zha, Lu Gan, Qianmu Li
{"title":"DS-TFSN-Based Vehicle Travel Time Prediction Method for Digital Twin System of Freeways","authors":"Weibin Zhang, Huazhu Zha, Lu Gan, Qianmu Li","doi":"10.1109/tits.2024.3451714","DOIUrl":"https://doi.org/10.1109/tits.2024.3451714","url":null,"abstract":"","PeriodicalId":13416,"journal":{"name":"IEEE Transactions on Intelligent Transportation Systems","volume":"40 1","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142264171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：