A deep learning-based distracted driving detection solution implemented on embedded system

This disclosure, in general, relates to enabling a between the one or more Internet of Things (IoT) device group communication. In particular, various heterogeneous IoT device that may need to interact with each other in different ways, may be organized into IoT device group to support the efficient interaction between IoT device. For example, pre-defined IoT device groups may be formed of a predetermined IoT devices that perform similar activities, certain IoT devices add for a given context-be dynamically allocated to the lump IoT device group It may be (e.g., ad-hoc IoT device group may comprise IoT device that can implement a desired function and thus may be dynamically formed in the to implement the desired functionality). In addition, IoT group may communicate in a hierarchical, wherein the message may support efficient communication between different IoT group is exchanged between the IoT group owner or ranking member.

Establishing groups of internet of things (iot) devices and enabling communication among the groups of iot devices

Road accidents are one of the biggest problems in the world, in which many precious lives have been lost. This work proposes a road vehicle accident detection system with location alerts to rescue accident victims. The main hardware modules include the MPU9250, a 9 degree-of-freedom (9-DoF) micro electro mechanical system (MEMS) based inertial measurement unit (IMU), Arduino nano with ESP8266 microcontroller and GPS. The IMU’s three axis accelerometer, gyroscope and magnetometer data are programmed to determine the orientation and position of the vehicle. A Wi-Fi-based communication is established using ESP8266 to send data received from sensor units to Google Firebase cloud servers in real-time. The performance of the developed device has been evaluated using a laboratory setup and also in real-time driving scenarios. The developed sensor module performs well on accident detection and emergency alert generation, which can be used in vehicles to save many lives in the event of an accident through its automatic alert service.

IoT-Based Road Accident Rescue System Implementation for Smart City Applications

This paper presents an efficient simulation model using OMNET++ and SUMO to establish inter-vehicle communication for proper traffic management and road safety in case of road traffic accidents. The generation of network and traffic simulators, i.e., OMNET++ and SUMO respectively, are connected using "VEINS", an open source framework for running network simulation of vehicles. The output is intended to allow different vehicles to communicate with each other in general and on a manually generated map by data packet transfer on specifically triggered events. The packet transfer has been analyzed and the results obtained have shown proper data communication between the nodes of the vehicles indicating that immediate care can be taken to rescue the accident victims. The novelty reflects the performance of using dedicated short-range communication (DSRC) of approximately 300 meters for efficient management of power requirements. Reducing the transmit power to 9mW also makes the system power efficient and the use of the UDP protocol increases the transmission speed, resulting in higher data rates. In the future, the developed model can be adapted according to the requirements of the actions to be taken on different incidents. Different beacons can be added to work according to different models.

An Efficient Inter-Vehicle Communication Framework on Road Traffic Accident Detection using OMNET++ and SUMO

The extension of transportation networks has sped up the speed of our lives. Car crashes cause a significant measure of death, property harm, and inestimable time, which is a significant worldwide medical condition. It is viewed as one of the essential executioners in the advanced world. The plan of a savvy mishap identification, area following, and cautioning framework that can detect mishaps as they happen is shrouded in this article. To find the mishap's definite area, a GPS device is utilized. The Worldwide Framework for Versatile (GSM) module sends a notification message that remembers a connection to the mishap's area for a Google guide to the nearby emergency clinic and police control focus. They can do whatever it may take to facilitate the salvage exertion by visiting the connection to find where the mishap is in the area.

Implementation of Accident Detection and Reporting System Using IOT

Road traffic crashes are the major risk factor in everyone’s day-to-day life due to the fast increase in the number of vehicles. It is now a more challenging task to deal with existing traffic systems with massive traffic. In the current scenario, the Internet of things (IoT) based solutions will make the transport system more intelligent. This solution will help to reduce congestion and increase the safety of everyday people. Fast reporting of driving abnormalities1 will help in preventing the life of the person involved in traffic crashes. Traffic crashes can be addressed using sensors, audio, and video-based analysis. The work presented here is a fast information sharing system with the date and time of the event, the detailed geographical location with Google map URL, speed, and the path traced using global positioning system (GPS) data. The fast reporting system uses various sensors to collect the vehicle data and notifies the driving abnormality, whenever any abnormal situation occurs. It will help the rescue team, insurance people, or relatives for easy navigation to the desired location. The limitations of communication bandwidth requirement and the powerful central processor may reduce detection performance; however, fast reporting could save many precious lives from danger.1 In this article, the abnormality is the behavior of the vehicle when it meets an accident while on the move.

http://dspace.nitrkl.ac.in:8080/dspace/bitstream/2080/3544/1/SahooG_Hydcon2020.pdf

S Aloka Patro

Papers

An IoT-Based Intimation and Path Tracing of a Vehicle Involved in Road Traffic Crashes

A Two-Tier Machine Learning Framework for Risk Assessment in Drivers with Cardiovascular Disorders