The Application Study of the Assisted-Driving System Based on Internet of Things System

Xinyu ZhangThe Frontiers of Society Science and Technology 2021 · DOI

TL;DR

This paper proposes an Assisted Driving System (ADS) leveraging Internet of Things (IoT) concepts to enhance road safety and prevent collisions. The system integrates sensors, a central controller, and actuators to enable automated emergency braking and steering adjustments, interacting with both vehicles and infrastructure.

What Problem Does This Paper Address?

Motor vehicle collisions remain a significant global safety concern, leading to numerous fatalities and injuries annually. While fully autonomous vehicles are still some way off, there's a pressing need for advanced driver-assistance systems that can actively intervene to prevent accidents. This paper addresses the challenge of reducing traffic accidents by developing an IoT-based Assisted Driving System (ADS) that can predict and react to potential collisions in various scenarios, including interactions with pedestrians, cyclists, other vehicles, and infrastructure.

Methods

The proposed ADS is structured into four subsystems for comprehensive accident prediction:

  • Sensors: A suite of sensors including LIDAR, Vehicle Speed Sensors (VSS), GPS, cameras (including a 360° view camera), and RFID sensors are employed to gather real-time data on the vehicle's speed, location, surrounding environment, and infrastructure.
  • Central Controller: This component processes the data from sensors, analyzes potential collision risks, and formulates appropriate commands. It acts as a data translator, integrating information from various sources to make informed decisions.
  • Actuators: These are the control mechanisms that execute the commands from the central controller. They include systems for accelerator control, brake control, and steering control, enabling the vehicle to adjust its speed and direction.
  • Internet of Things (IoT) Integration: The system leverages IoT concepts for communication and data exchange, particularly through Vehicle-to-Everything (V2X) communication, including Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I).

Specific interaction modules are detailed:

  • Vehicle-to-Infrastructure (V2I): This involves interactions with traffic lights (using RFID tags) and stop signs (recognized by cameras). It also addresses roadworks and urban infrastructure hazards by obtaining information from cloud services and local maps.
  • Vehicle-to-Non-Vehicle: This covers interactions with pedestrians and cyclists. For pedestrians, radar sensors and mobile app notifications are used. For cyclists, a Simplified System Device (SSD) with radar sensors is proposed for warning.
  • Vehicle-to-Vehicle (V2V): This subsystem focuses on analyzing the behavior of other vehicles using sensor data and V2V communication (via VANETs) to predict and avoid collisions.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Holistic Approach: The ADS is designed to be comprehensive, considering not just other vehicles but also pedestrians, cyclists, and infrastructure as potential collision factors.
  • Layered Architecture: The system's layered structure (Sensors, Embedded System, Actuators) provides a clear framework for data processing and control.
  • V2X Communication is Crucial: The paper emphasizes the importance of V2X communication (V2V and V2I) for obtaining real-time information about the environment and other road users, which is vital for accurate prediction and intervention.
  • Simplified Devices for Non-Motorized Users: The proposal of a Simplified System Device (SSD) for cyclists highlights a practical approach to extending safety features to vulnerable road users who may not have sophisticated onboard systems.
  • Beyond Collision Avoidance: The paper suggests that the data collected by the ADS could have broader applications, such as monitoring driver habits for insurance purposes or aiding law enforcement.

Key Results

The paper outlines the functional capabilities of the proposed ADS rather than presenting quantitative experimental results. The key "results" are the conceptual design and the proposed mechanisms for:

  • Automated Emergency Braking (AEB): The system can recognize potential collisions and initiate braking.
  • Steering Control: The system can adjust steering to avoid obstacles.
  • Interaction with Infrastructure: The ADS can interpret traffic lights and stop signs.
  • Warning Systems: Alerts are provided to drivers and, in some cases, to pedestrians and cyclists.
  • Adaptation to Conditions: The system aims to function in various conditions, including extreme weather.

Limitations and Future Work

  • Lack of Empirical Validation: The paper presents a conceptual design and does not include experimental results or performance metrics (e.g., accuracy, reaction times, collision avoidance rates).
  • Reliance on Cloud Data: For roadworks and urban infrastructure, the system relies on cloud data, which might be subject to timeliness and accuracy issues.
  • Complexity of Real-World Scenarios: While the paper outlines various subsystems, the complexity of integrating and ensuring the seamless operation of all these components in diverse real-world scenarios remains a significant challenge.
  • Future Work: The authors believe the proposed system has "great application prospects," implying that further development and testing are envisioned. The paper also hints at future work in areas like machine learning for data transmission reduction and broader applications beyond collision avoidance.

Summary updated

Abstract

In view of the instability and high-risked motor vehicle collision, this work proposes an Assisted Driving System (ADS) based on Internet of Things (IoT) concepts to prevent traffic accident. In detail, ADS utilizes Automated Emergency Braking (AEB) and embedded actuators in the steering wheel to control the vehicle in an emergency situation. AEB consists of four different subsystems that allow the vehicle to recognize potential collision and need to full-stop in interaction with pedestrian, biker, another vehicle, and infrastructure. Pedestrians and bikers need an application or Simplified System Device (SSD) to receive feedback from the interaction. Radar sensors, front-camera, RFID sensor and Speedometer also allow the ADS to operate on a single-vehicle circumstances. It is believed that the proposed system of this work has great application prospects.
BibTeX
@inproceedings{zhang2021application,
  author = {Xinyu Zhang},
  title = {{The Application Study of the Assisted-Driving System Based on Internet of Things System}},
  booktitle = {The Frontiers of Society Science and Technology},
  year = {2021},
  doi = {10.25236/fsst.2021.030205},
}
Figure