Emergency Scenarios Research Articles

Real-time human-AI collaboration through scalable cloud platforms for emergency response

Real-time human-AI collaboration is revolutionizing emergency response, yet challenges remain in achieving seamless interaction at scale. This article explores an innovative approach leveraging scalable cloud platforms to enable effective collaboration between human responders and AI systems during critical incidents. Integrating cloud-native solutions ensures real-time data processing, rapid decision support, and dynamic adaptation to evolving scenarios. Key features include adaptive load balancing to accommodate fluctuating data streams, AI-driven predictive analytics for preemptive action, and intelligent communication channels to enhance coordination among responders. The proposed architecture minimizes latency, optimizes resource allocation, and maintains service continuity, even under extreme conditions. The framework addresses data security, scalability, and compliance challenges to offer a robust, reliable solution for time-sensitive operations. Implementation results demonstrate significant improvements in response times, incident handling capacity, and resource utilization across multiple real-world emergency scenarios.

Optimization of Distributed UAV Swarm Placement for Target Localization Using Multiple Heuristic Algorithms Based on Compressive Sensing

Collaborative UAV swarms are increasingly deployed as temporary base stations in emergency situations to relay communications. This paper designs intelligent optimization algorithms, which minimizing mutual coherence within a region of interest (ROI) aims to solving challenge in these scenarios is optimizing the UAVs' locations to avoid mutual signal interference and ensure high communication quality. Minimizing mutual coherence reduces the likelihood of signal interference between UAVs. We explore various optimize algorithum, including Heuristic Search (HS) and Ant Colony Optimization (ACO) and so on. The results provide insights into each algorithm's performance in dynamic environments, helping to identify the most suitable approaches for UAV deployment in emergency scenarios. This study contributes to the development of efficient UAV deployment strategies, enhancing the reliability of UAV-based communication systems during critical events.

Using artificial intelligence based language interpretation in non-urgent paediatric emergency consultations: a clinical performance test and legal evaluation

ObjectiveTo evaluate the accuracy of Google Translate (GT) in translating low-acuity paediatric emergency consultations involving respiratory symptoms and fever, and to examine legal and policy implications of using AI-based language interpretation in healthcare.MethodsBased on the methodology used for conducting language performance testing routinely at the Interpreter Services Department of the Hospital for Sick Children, clinical performance testing was completed using a paediatric emergency scenario (child with respiratory illness and fever) on five languages: Spanish, French, Urdu, Arabic, and Mandarin. The study focused on GT's translation accuracy and a legal and policy evaluation regarding AI-based interpretation in healthcare was conducted by legal scholars.ResultsGT demonstrated strong translation performance, with accuracy rates from 83.5% in Urdu to 95.4% in French. Challenges included dialect sensitivity and pronoun misinterpretations. Legal evaluation indicated inconsistent access to language interpretation services across healthcare jurisdictions and potential risks involving data privacy, consent, and malpractice when using AI-based translation tools.ConclusionsGoogle Translate can effectively support communication in specific non-critical paediatric emergency scenarios. However, its use necessitates careful monitoring, understanding of its limitations, and attention to dialect and literal translation risks along with equity considerations. Establishing legal and policy frameworks for language interpretation in healthcare is crucial, alongside addressing funding and data security concerns, to optimize the use of AI-based translation tools in healthcare contexts.

"Comparative analysis of large language models against the NHS 111 online triaging for emergency ophthalmology".

This study presents a comprehensive evaluation of the performance of various large language models in generating responses for ophthalmology emergencies and compares their accuracy with the established United Kingdom's National Health Service 111 online system. We included 21 ophthalmology-related emergency scenario questions from the NHS 111 triaging algorithm. These questions were based on four different ophthalmology emergency themes as laid out in the NHS 111 algorithm. Responses generated from NHS 111 online, were compared to different LLM-chatbots responses to determine the accuracy of LLM responses. We included a range of models including ChatGPT-3.5, Google Bard, Bing Chat, and ChatGPT-4.0. The accuracy of each LLM-chatbot response was compared against the NHS 111 Triage using a two-prompt strategy. Answers were graded as following: -2 graded as "Very poor", -1 as "Poor", O as "No response", 1 as "Good", 2 as "Very good" and 3 graded as "Excellent". Overall LLMs' attained a good accuracy in this study compared against the NHS 111 responses. The score of ≥1 graded as "Good" was achieved by 93% responses of all LLMs. This refers to at least part of this answer having correct information as well as absence of any wrong information. There was no marked difference and very similar results seen overall on both prompts. The high accuracy and safety observed in LLM responses support their potential as effective tools for providing timely information and guidance to patients. LLMs hold promise in enhancing patient care and healthcare accessibility in digital age.

On-site visual quantification of alkaline phosphatase activity in cells using a smartphone-based approach.

Alkaline phosphatase (ALP) is a critical biomarker associated with various physiological and pathological processes, making its detection essential for disease diagnosis and biomedical research. In this study, we developed a novel, simple, and portable visual quantification method for ALP activity in cells using an efficient Cu0.9Zn0.1S nanomaterial with peroxidase-like properties, integrated into a smartphone-based platform for enhanced usability. The Cu0.9Zn0.1S nanomaterial catalyzes the breakdown of H₂O₂, generating ·OH radicals that oxidize the colorless substrate TMB into blue oxTMB, which is subsequently reduced back to TMB by ascorbic acid (AA). We employed an indirect colorimetric approach for ALP detection, leveraging ALP's ability to catalyze the dephosphorylation of l-ascorbic acid-2-phosphate (AAP) to produce AA. This method enabled highly sensitive and precise quantification of ALP with a detection limit of 0.47 mU/L and a linear range from 0.001 to 100 U/L. The established smartphone platform not only facilitated real-time data processing but also transformed the detection system into a portable and accessible laboratory. The method was successfully applied to detect ALP in various cancer cell lines, with the highest levels found in HepG2 cells. Moreover, the results were consistent with those obtained using traditional kit-based methods, highlighting the accuracy and reliability of our approach. The system was also applied to evaluate ALP inhibitors, demonstrating its versatility in biomedical applications. This innovative, cost-effective, and highly sensitive colorimetric sensing strategy offers immense potential for clinical diagnostics, cancer research, and inhibitor screening, particularly in on-site, resource-limited, or emergency scenarios.

Hemoglobin-loaded ZIF-8 Nanoparticles Equipped with PEGylated Metal-phenolic Network Coatings: An Oxygen Carrier with Antioxidant and Stealth Properties

Hemoglobin-based oxygen carriers (HBOCs) offer a promising alternative to conventional blood transfusions in emergency scenarios. However, achieving optimal stability, functionality, and biocompatibility in HBOCs remains a significant challenge. In this...

Emergency scenario modeling for the analysis of dynamic risks in business parks

Emergency scenario modeling for the analysis of dynamic risks in business parks

AI-Driven Progressive Web Application for Enhancing Emergency Healthcare Decision-Making

The "Emergency Lifeline: Instant Hospital Info for Critical Moments" mobile application aims to revolutionize emergency healthcare by providing real-time hospital information, medical guidance, and optimized routing during critical situations. The application integrates advanced technologies such as Artificial Intelligence (AI), Geolocation, Telemedicine, and Multilingual Support to offer a comprehensive emergency solution. By leveraging AI for symptom analysis and hospital recommendations, GPS for real- time location tracking and optimized routing, and telemedicine for remote consultations, the app ensures that users can access medical assistance quickly and effectively. The inclusion of multilingual support, along with voice recognition and text-to- speech capabilities, ensures accessibility for a diverse user base, including those with disabilities or literacy challenges. This innovative approach enables users to make informed decisions, minimize response times, and improve healthcare outcomes in emergency scenarios. The system's design focuses on user experience, data security, and collaboration with healthcare providers to ensure accurate, up-to-date information. Overall, the app aims to enhance the efficiency and accessibility of emergency healthcare, making it a vital tool for individuals facing critical health situations. Keyword - Emergency healthcare, mobile application, Artificial Intelligence (AI), Geolocation, Telemedicine, Multilingual support, Real-time hospital information, Symptom analysis, Routing optimization, Text-to-speech, Voice recognition, Healthcare accessibility, Critical moments, Medical guidance, User experience, Data security.

INFORMATION AND SOFTWARE OF VIRTUAL REALITY TRAINING APPLICATION «NATURAL GAS DRYER PLANT»

Abstract. Traditional training methods for natural gas dehydration plant operators have several limitations. Theoretical training may need to provide more insight into real-world working conditions, while practical training can be dangerous or require significant financial investment. Virtual reality (VR) technology offers a potential solution to this challenge by enabling the replication of the natural gas dehydration plant experience in a safe, immersive digital environment. This research project has developed a software application that simulates the operation of a natural gas dehydration plant using VR technology. The article describes the world background and possible alternatives for the project's hardware and software implementation. Finally, the virtual environment was created using the Unity game engine, which incorporates Open VR and VRTK plugins to enable user interaction and navigation. The application allows users to explore the plant freely, interact with equipment and tools, perform standard operating procedures, and respond to emergency scenarios. The project has several scenarios controlled by the tutor that need to be solved by the student. All scenarios involve visual effects to increase immersion. A multi-user mode and voice chat functionality have also been implemented to facilitate collaborative learning. The development of this VR training application addresses the need for practical, hands-on learning in the natural gas industry, where actual equipment can be prohibitively expensive or hazardous. By providing a realistic, interactive virtual environment, the application aims to enhance the skills and understanding of natural gas dehydration plant operators, improving safety and operational efficiency. The application was developed on request and in close cooperation with the Department of Chemical Engineering at Sumy State University. The findings from this research can serve as a basis for further development of VR applications in various industrial and educational domains.

The Clinical Efficacy and Applications of Portable Imaging Devices in Emergency Medical Settings: A Comprehensive Review

Background: The integration of portable imaging devices in emergency medical settings has gained attention due to their potential to enhance diagnostic efficiency and facilitate timely patient care. Initial advancements in mobile imaging technologies under the Digital Imaging and Communications in Medicine (DICOM) standard date back to 2003, yet their clinical adoption remains limited. Methods: This review synthesizes findings from various studies sourced from prominent medical databases, including PubMed, Wiley Online Library, Web of Science, and Scopus. Key search phrases related to mobile DICOM viewers were employed to gather relevant literature assessing the diagnostic efficacy of mobile devices in interpreting radiological images in emergency scenarios. Results: Several studies highlighted that mobile device, including tablets and smartphones, can effectively display and interpret radiological images, offering diagnostic capabilities comparable to traditional PACS workstations. For instance, mobile applications significantly reduced consultation times and enhanced accessibility to imaging data. However, the European Society of Radiology cautioned against their use for primary interpretations, recommending them instead for supplementary opinions or bedside evaluations. Conclusion: The findings underscore the feasibility of utilizing portable imaging devices in emergency settings to improve access to diagnostic imaging and expedite clinical decision-making. Despite their advantages, ongoing concerns regarding image quality, usability, and the need for robust IT infrastructure must be addressed to facilitate broader adoption in clinical practice.

Establishment of a Low-Cost Reinforced Concrete Prefabricated Family Housing Construction System

Objective: The objective of this study is to investigate the feasibility of an affordable and rapidly deployable prefabricated housing system adapted to urban contexts and emergency scenarios. The aim is to design an integrated system for the production, construction, and assembly of prefabricated reinforced concrete housing, offering accessible, efficient, and flexible housing solutions. Theoretical Framework: The research is grounded in key concepts and theories such as prefabrication in construction, modularity, environmental sustainability, housing resilience, advanced concrete technology, and project management on-site, providing a solid basis for understanding its context. Methodology: The study employs a comparative method, analyzing six representative housing units with built areas of 25.83 m², 33.39 m², 39.06 m², 42.21 m², 52.29 m², and 59.85 m². Data collection occurred in three stages: (1) search and review of experiences, (2) design and development of basic housing models, production, and assembly, (3) results analysis to evaluate technical and economic feasibility. Results and Discussion: The results confirm the feasibility of designing six models of basic family housing that comply with Peruvian standards and are accessible to low-income families. A production process for prefabricated elements is proposed, adaptable to any location and operable with minimal equipment. Among the evaluated models, VUF 04 stands out for its lowest cost, estimated at $311.43 USD per square meter of built area. Research Implications: The implications of this research are broad and significant, highlighting its potential to transform the design, production, and assembly of housing in urban and emergency contexts. By addressing multiple aspects, this study establishes a solid foundation for the implementation of sustainable, accessible, and efficient housing solutions. Originality/Value: The originality of this study lies in its decentralized and simplified production approach, carried out on-site with minimal equipment. Its value resides in democratizing access to affordable, efficient, and sustainable housing, providing an alternative model to centralized systems and laying the groundwork for future research on accessible and resilient construction.

On the optimal integration of legacy and SDWN communication nodes in emergency scenarios

AbstractSpontaneous networks may need to establish connections with already-in-place devices and their resources, irrespectively of their capabilities, for example, in emergency scenarios. Interoperability and integration are a key issue when combining multiple wireless networks and heterogeneous devices. To address this issue, this article extends the SDN concept, and presents a new hybrid network architecture that combines legacy devices in a wireless data plane, therefore increasing the range, and making the network more efficient, dynamic and programmable. In this article, we propose an optimized adaptive routing approach in a network architecture that joins SDN-capable devices with legacy devices for emergency scenarios. This solution takes into consideration the impact that a flow can have in the network, therefore minimizing its channel interference, and to a certain degree, balancing the load in the network’s devices. In this work, the importance of the network and service parameters is optimized through machine learning for each flow’s data rate, and then feed into the routing approach to determine the best hybrid legacy and SDN-based path. Results show that the new architecture can successfully integrate the legacy devices in the wireless software defined data plane, and utilize them to increase the performance. The combination of delivery ratio and delay are always relevant to every data rate. The link stability is more relevant for lower data rates, while the break probability is more relevant for higher data rates. Moreover, the number of hops and possible interference combined have a high importance in every data rates, and the nodes’ speed knowledge can have a good impact on the network performance.

A multi-criteria approach to egress location selection in buildings: integrating economic, technical and social factors

PurposeThis paper aims to introduce an innovative decision-making framework designed to optimize egress location selection in buildings, significantly enhancing safety during emergencies. By integrating social, economic and technical criteria through the fuzzy analytic hierarchy process (F-AHP), this framework effectively manages the uncertainties inherent in emergency scenarios, providing a robust tool for safeguarding lives and property.Design/methodology/approachThe proposed methodology employs the F-AHP to systematically integrate and evaluate social, economic and technical criteria for egress location selection. A comprehensive case study conducted on a one-story school building demonstrates this approach’s practical application and effectiveness.FindingsThe analysis reveals that economic criteria are the most significant in determining the optimal egress location, constituting 61.9% of the overall weight, followed by technical criteria at 26.8% and social criteria at 11.3%. The findings highlight the practical application and effectiveness of the proposed approach in addressing uncertainties and optimizing egress location selection. Integrating social, economic and technical considerations gives decision-makers a robust tool to optimize safety, cost-effectiveness and evacuation efficiency.Practical implicationsThe findings underscore the practical benefits and effectiveness of the proposed approach in managing uncertainties and improving the selection of optimal egress locations. Integrating social, economic and technical considerations gives decision-makers a robust tool to optimize safety, cost-effectiveness and evacuation efficiency. This approach equips practitioners with a comprehensive framework for enhancing emergency preparedness and resilience in various building scenarios.Originality/valueThis research introduces a novel framework for egress location selection that addresses the limitations of existing studies by integrating multiple criteria – social, economic and technical – into a cohesive decision-making process. Using the F-AHP in a case study illustrates the framework’s practicality and adaptability, providing a dynamic solution that enhances the overall effectiveness of emergency plans. By prioritizing safety, cost-effectiveness and efficiency, this framework ensures that emergency plans are robust and adaptable, thereby significantly enhancing resilience in the face of emergencies.

Evaluating Pilot Mental Workload Using fNIRS-Based Functional Connectivity Features with a Deep Residual Shrinkage Network Under Emergency Flight Scenarios

Excessive mental workload can lead to less remaining resources for pilots to perform concurrent tasks during emergency flights, affecting aviation safety. Based on a flight simulator, this study investigated 25 cadet pilots using functional near-infrared spectroscopy (fNIRS) and subjective ratings to assess their mental workload under three subtasks with different equipment failures. fNIRS data included oxyhemoglobin, deoxyhemoglobin, and total hemoglobin signals, yielding 10545 functional connectivity (FC) features from four brain regions: prefrontal, right motor, left motor, and occipital cortexes. A deep residual shrinkage network classified mental workload levels, outperforming convolutional neural network and random forest models with 89.58% accuracy after feature selection employing an interpretable machine learning algorithm. The results suggest that brain FC from three hemoglobin signals could be used to differentiate the three different levels of pilot mental workload. This study could contribute to improving pilot training and supporting the development of pilots’ competencies during emergency scenarios.

Blockchain empowered dynamic access control for secure data sharing in collaborative emergency management

In the realm of emergency management involving multi-party collaboration, securely sharing data among multiple entities poses a significant challenge. Traditional centralized data platforms struggle with fine-grained access control for data from different sources and are susceptible to single-point failures and risks of forged authorization. This paper delves into the prospective application of blockchain technology for decentralized data governance. We propose a dynamic access control technology for multi-party collaborative emergency response, which combines blockchain and attribute-based access control technologies to enable dynamic access control in multi-party data sharing scenarios. A data sharing system for collaborative emergency management built with Hyperledger Fabric has been developed and subjected to a series of evaluation tests. The experimental results reveal that our blockchain-based system not only significantly enhances data security and privacy by leveraging decentralized control but also improves the efficiency and reliability of data sharing in emergency situations. Specifically, our findings demonstrate the system's ability to dynamically adapt access control policies based on changing emergency scenarios, effectively mitigating risks associated with single-point failures and unauthorized access. These outcomes underscore the viability and reliability of our approach, providing a robust framework for secure, scalable, and flexible data sharing in the demanding field of emergency management.

Intelligent Decision Support Systems for Optimizing Medical Emergency Responses

Now-a-days, in the realm of an emergency medical services-EMS, the swift and an accurate decision-making is a critical for an ensuring of timely responses and all the optimal patient outcome results. With a heavy advent of an advanced techniques and these proliferation of the data sources in the healthcare, there is a high-level growing opportunity of an Intelligent decision supporting systems-IDSS to enhance an efficiency and an effectiveness of the medical emergency responses-MER widely. This study paper explores the role of an intelligent-DDS in the optimization of the medical emergency responses through analyzing various aspects factors like of the real-time analytic, data integration, and the decision supporting-algorithms. First of all, the integration of the diverse data-sources likewise patients healthcare records, some historical incidents, and the geo-graphical information all forms the foundation of an effective IDSS. The amalgamation of the data streaming generated by an IDSS can be comprehensive awareness, enabling the res-ponders to make an informed-decision tailored to any specific needs in each emergency scenarios. Then, the real-time analytic plays a pivotal role in the interpretations and processing of an incoming data-sets to detect the patterns, trends, and the anomalies. With the help of these techniques like machine-learning as well predictive-modeling, an IDSS can simply anticipate the potential emergency, pro-actively allocated the resources, and optimized response routes too in an-order-to minimize the time-to treatments of the patients globally. Furthermore, decision-support algorithms embedded within IDSS provide respondents with an actionable insights and recommendations based-on analysis of available data-sets. These powerful IDSS algorithms considered factors like severity of medical conditions, proximity to the health-care facilities, and as well the availability of specialized equipment to guide respondent in prioritizing tasks and allocating resources efficiently and enhancing over-all systems resilience. [1-3]

INTELLIGENT MONITORING FOR DETECTIONS AND ESTIMATING RADIOLOGICAL PLUME DISPERSION IN AQUATIC ENVIRONMENTS: RIVERS AND SEAS

This work continues the presentation INTELLIGENT MONITORING FOR RADIOLOGICAL PLUME DISPERSION ESTIMATION - INTELLIGENT MONITORING FOR RADIOLOGICAL PLUM DISPERSION ESTIMATION, developed as part of the event hosted by ABDAN at the third edition of the Nuclear Trade & Technology Exchange – NT2E, Brazilian Nuclear Olympics (ONB), Hackapower. The primary objective was to develop a system for monitoring radiological plumes in aquatic environments, using a model based on Gaussian Process Regression (GPR) and Particle Swarm Optimization (PSO). The proposed system aims to estimate radiation dispersion in scenarios where plant systems are unavailable, similar to the Fukushima accident. Given this context, it was necessary to create mobile devices with radiation detection capabilities to provide accurate data on dose rate distribution. The model was developed to predict dose distribution using simulated data from a hypothetical accident and involved the use of LoRaWAN networks for drone communication and a firefly algorithm for signaling areas with different radiation levels. The implementation of GPR utilized the ScikitLearn library, while PSO was applied through the pyswarm library, focusing on optimization based on information entropy. Results indicated that the model successfully reconstructed the dose rate profile with an estimate close to the actual values, although data non-uniformity may have impacted accuracy. The use of drones for data collection proved innovative and effective, enabling real-time analysis and offering a robust solution for radiological monitoring in emergency scenarios. Analysis of radiation permeability variation in different aquatic environments highlighted the importance of adjusting measurements according to water density and composition. In conclusion, the work achieved its goal by developing an intelligent system for radiation dispersion estimation, and future work should explore new scenarios and dynamics to enhance model accuracy in real radiological emergency situations.

Remotely geolocating static events observed by citizens using data collected by mobile devices

The increasing use of smartphones has led to a surge in crowdsourcing initiatives, where citizens easily collect and upload information using advanced sensors, leveraging the collective efforts of the crowd. However, these devices face accuracy issues that must be addressed before they can be used effectively in certain applications. While most research has focused on GNSS-based positioning errors, compass-based orientation errors have received far less attention. This paper presents a novel method for determining the geolocation of static events by combining contributions from multiple volunteers located around a target event. Instead of straight lines, the method applies funnel-shaped regions to represent the azimuths toward the target location and encompasses various phases to address errors stemming from measured azimuths. These compass-based orientation errors have attracted less attention compared to GNSS-based positioning errors. The approach mitigates the uncertainty in azimuth measurements, producing a set of regions that contain the target location with different levels of confidence. Its reliability was tested in three case studies with known target locations, using metrics such as area and compactness. The results were promising, indicating that the developed approach is particularly useful in applications that do not require a high degree of accuracy (common in many crowdsourcing projects), such as the geolocation of forest fire ignitions − once in the vicinity they can be easily spotted. Implementing this approach within a system that collates citizens’ contributions can furnish invaluable information to authorities, empowering prompt action during emergency scenarios.

Evaluating the Utility of ChatGPT in Diagnosing and Managing Maxillofacial Trauma.

Maxillofacial trauma is a significant concern in emergency departments (EDs) due to its high prevalence and the complexity of its management. However, many ED physicians lack specialized training and confidence in handling these cases, leading to a high rate of facial trauma referrals and increased stress on consult services. Recent advancements in artificial intelligence, particularly in large language models such as ChatGPT, have shown potential in aiding clinical decision-making. This study specifically examines the efficacy of ChatGPT in diagnosing and managing maxillofacial trauma. Ten clinical vignettes describing common facial trauma scenarios were presented to a group of plastic surgery residents from a tertiary care center and to ChatGPT. The chatbot and residents were asked to provide their diagnosis, ED management, and definitive management for each scenario. Responses were scored by attending plastic surgeons who were blinded to the response source. The study included 13 resident and ChatGPT responses. The mean total scores were similar between residents and ChatGPT (23.23 versus 22.77, P > 0.05). ChatGPT outperformed residents in diagnostic accuracy (9.85 versus 8.54, P < 0.001) but underperformed in definitive management (8.35 versus 6.35, P < 0.001). There was no significant difference in ED management scores between ChatGPT and the residents. ChatGPT demonstrated high accuracy in diagnosing maxillofacial trauma. However, its ability to suggest appropriate ED management and definitive treatment plans was limited. These findings suggest that while ChatGPT may serve as a valuable diagnostic tool in ED settings, further advancements are necessary before it can reliably contribute to treatment planning in emergent maxillofacial clinical scenarios.

Performance evaluation for Q-learning based anycast routing protocol in unmanned aerial vehicle networks with multiple base stations

Unmanned Aerial Vehicle (UAV) networks can be used for data transmission in emergency scenarios, relaying data from ground users to base stations (BSs). While UAV networks collaborating with multi-BSs can significantly enhance performance, existing UAV routing protocols predominantly focus on unicast routing and often neglect critical aspects such as base station discovery. In addition, the high mobility of UAVs and rapid changes in network topology also pose great challenges for existing multi-BS routing protocols to maintain efficient data transmission. Aiming at the above problems, this paper abstracts the routing of multi-base station UAV networks as anycast routing for dynamic networks and proposes a distributed anycast routing protocol called QARP to improve the data transmission performance. In QARP, base stations can be discovered automatically and parameters of Q-learning are dynamically adjusted to improve the efficiency of data transmission. The Link Duration Estimation is used to influence routing decision and dynamically adjust the hello message interval. A multiple base stations transmission value function is designed to indicate the performance of data transmission and is used to calculate the reward and update Q-table. The experimental results show that the QARP proposed in this paper outperforms existing multi-BS routing and Q-learning based routing protocols in terms of delay, packet delivery ratio and throughput in single base station and multiple base stations scenarios.