Driving Assistance System for Ambulances to Minimise the Vibrations in Patient Cabin

Abstract: The ambulance service is the main transport for diseased or injured people which suffers the same acceleration forces as regular vehicles. These accelerations, caused by the movement of the vehicle, impact the performance of tasks executed by sanitary personnel, which can affect patient survival or recovery time. In this paper, we have trained, validated, and tested a system to assess driving in ambulance services. The proposed system is composed of a sensor node which measures the vehicle vibrations using an accelerometer. It also includes a GPS sensor, a battery, a display, and a speaker. When two possible routes reach the same destination point, the system compares the two routes based on previously classified data and calculates an index and a score. Thus, the index balances the possible routes in terms of time to reach the destination and the vibrations suffered in the patient cabin to recommend the route that minimises those vibrations. Three datasets are used to train, validate, and test the system. Based on an Artificial Neural network (ANN), the classification model is trained with tagged data classified as low, medium, and high vibrations, and 97% accuracy is achieved. Then, the obtained model is validated using data from three routes of another region. Finally, the system is tested in two new scenarios with two possible routes to reach the destination. The results indicate that the route with less vibration is preferred when there are low time differences (less than 6%) between the two possible routes. Nonetheless, with the current weighting factors, the shortest route is preferred when time differences between routes are higher than 20%, regardless of the higher vibrations in the shortest route.

• The paper presents a sensor-driven route recommendation system that uses ANN and time-buffered standard deviation features to classify road conditions with 97% accuracy.
• It introduces index and score metrics to balance travel time and cabin vibration, optimizing patient comfort during emergency transport.
• Real-world tests confirm the system’s robust performance in diverse urban and interurban conditions while alerting drivers in high-vibration zones.

Driving Assistance System for Ambulances to Minimise Vibrations in the Patient Cabin

Research Motivation and Context

Ambulance transport imposes unique physiological risks on patients due to cabin vibrations and accelerations, especially during traversal across heterogeneous urban and interurban infrastructure. Prior literature emphasizes mechanical and suspension-based solutions for vibration attenuation, but few address post-patient pickup mobility optimization for the evacuation phase. Existing route-planning frameworks for emergency vehicles often neglect patient comfort by focusing solely on minimization of response times or spatial resource allocation. The present work addresses these deficiencies by proposing a route recommendation system informed directly by vibration data, thereby extending operational comfort and facilitating medical interventions en route.

System Architecture and Methodological Advances

The proposed driving assistance system integrates a low-cost sensor platform (accelerometer and GPS) with a Raspberry Pi-based node, cloud connectivity, and an interactive interface. Operation mode 1 is dedicated to data acquisition and tagging, whereas operation mode 2 delivers real-time route recommendations incorporating previously classified vibration profiles.

Route classification into mobility areas (A1: interurban-good pavement, A2: urban avenues/interurban-regular, A3: urban streets/interurban-bad pavement) is achieved via an artificial neural network (ANN). The ANN's architecture leverages velocity and acceleration inputs; empirical assessment determined that x-axis acceleration introduces noise, and y, z axes are optimal for distinguishing route-induced cabin vibrations. A key innovation is the introduction of time-buffered standard deviation features—specifically, a 29 s window—yielding substantial improvements in classification accuracy over traditional approaches utilizing raw signals.

Index and Scoring Metrics

To operationalize route recommendation, two quantitative measures are introduced:

• Index: Weighs route traversal duration across different mobility areas, penalizing A3 with the highest coefficient (2), followed by A2 (1.5), and A1 (1). The index offers a combined metric for route selection in emergencies where tradeoffs between vibration and time-to-destination are essential.
• Score: Normalizes the weighted mobility area durations by total trip time, facilitating comparative evaluation across alternative routes with divergent temporal profiles. The score ranges from 1 (optimal) to 2 (maximal vibration), serving as a clinical reference point for transport comfort optimization.

Empirical Results

ANN training on manually tagged datasets using y and z axis standard deviation (29 s buffer) achieved a maximum classification accuracy of 97%. Validation using three independent urban/interurban routes confirmed qualitative alignment between ANN-assigned mobility classes and real-route elements—especially in challenging segments such as pedestrian crossings and roundabouts. False positives in classification were minimized with increased buffer duration, indicating robustness against noise and transient environmental features.

Testing in real-world scenarios revealed critical decision thresholds:

• Time Difference < 6%: The system preferred routes with lower vibration (lower score/index), demonstrating that patient comfort supersedes minor increases in transport time.
• Time Difference > 20%: The shortest route is invariably recommended, even if vibration is higher, due to weighting factors prioritizing time in high-urgency contexts.
• The system’s output aligns with clinical priorities: minimizing exposure to rough segments when feasible, and alerting drivers before entering high-vibration mobility areas (A2/A3) via speaker interface.

Practical and Theoretical Implications

This research introduces an actionable framework for patient-centric route planning in ambulances, moving beyond mechanical vibration mitigation to dynamic, real-time mobility decision support. The demonstrated high ANN accuracy supports operational integration, where the presented system can augment existing clinical workflow by reducing patient discomfort and facilitating critical interventions (e.g., resuscitation, ventilation) during transport.

Practically, the index and score metrics, combined with ANN-based classification, are extensible across demographic strata and regional road typologies. The method offers scalability via cloud connectivity and compatibility with Google Maps, and is poised for integration into vehicular networks (VANETs). Theoretically, the approach suggests that intelligent feature engineering (time-buffered standard deviation) and axis selection can markedly improve classification performance in mobility-related AI applications, presenting a template for similar systems in other transport modalities.

Future Directions

Ongoing and future work is anticipated in several domains:

• Driver and vehicle-specific calibration: Evaluating inter-driver, wheel, and damper effects on vibration profiles and ANN classification.
• Comparative algorithmic performance: Benchmarking against kernel extreme learning machines, deep adversarial transfer networks, and competitive swarm optimizers.
• Interpretability: Addressing the transparency of ANN models for motion prediction and integrating explainability to route decision logic.
• Multimodal data integration: Extending feature sets to include traffic and weather data, and enabling direct deployment within VANET infrastructures.

Conclusion

The paper provides a technically rigorous, data-driven solution to ambulance route recommendation that directly incorporates patient-cabin vibration as a primary criterion. The combination of sensor-driven ANN classification, time-buffered features, and weighted index/score metrics demonstrates strong empirical performance and practical relevance. Future enhancements are anticipated to broaden system adaptability, interpretation, and integration within smart transport networks, with substantial implications for clinical and operational outcomes in emergency medicine transport systems.

(2604.16047)