border border

Intelligent Driving Circuit

Team members

Heng Yee Ying Beverly (EPD), Lim Yi Lin (EPD), Fong Jing Jie Edward (EPD), Edmund Ng Choon Eng (EPD), Tan Xiang Yun, Jason (ESD), Pang Bang Yong (ISTD)

Instructors:

Karthik Balkrishnan, Teo Tee Hui, Norman Lee Tiong Seng

Writing Instructors:

Grace Kong

Teaching Assistant:

Li Xueliang

Project Description:

An objective grading system that detects dangerous riding manoeuvres by learner motorcyclists and automatically assesses their competency using a combination of algorithms and Internet-of-Things solutions.


Problem: Competency Assessments

Competency assessments for learner motorcyclists are a subjective and manual process. As the most vulnerable vehicles on the road, accurate assessments of the rider’s competency are essential for ensuring the safety of all road users.


Solution: Intelligent Driving Circuit (IDC)

An automated grading system was designed to detect gaps in competency while riding through the circuit. Using a combination of sensors and algorithms, the system collects data of the motorcyclist’s riding behavior before classifying and detecting any faults.


Intelligent Driving Circuit


Features

1. Emergency Brake Station

Our system uses Time-of-Flight (ToF) sensors to assess ability to stop in emergencies.

  • Sensors aligned in array to maximise accuracy
  • Custom designed weatherproof box for durability
  • Easy to maintain and calibrate
  • 100% Accuracy within +/- 50mm tolerance


    • ebrake
    boxrenderfinal


    2. Wide or Narrow Turning

    Our system uses a curve fitting algorithm with IMU and GPS data of the motorcyclist to detect and penalize wide or narrow turning.

    sharpturn

    3. Wobbling

    Wobbling is detected using high pass filter and signal processing methods on the data collected from an IMU sensor fitted to the motorcycle.

    wobble

    4. Interactive General Riding

    Detects unsafe interactions such as following too closely behind vehicles or cutting into other lanes, by motion extrapolation via a kinematics model with bounding box collision detection.

    igr

    5. Abrupt Lane Change

    Abrupt lane changes are detected by feature engineering using statistical methods and thresholding large changes in yaw of the motorcycle.

    alcpng


    Full System Integration

    In addition to the features which we developed, we envision a graphical user interface (GUI) for a fully featured IDC to allow for an intuitive and better way to learn and improve on mistakes.

    untitled design 3
    untitled design 4

    Try our demo app:

    In summary, the IDC improves the competency assessment process by:


    1) Eliminating long wait time to tabulate results

    2) Reducing subjectivity in grading

    3) Ensuring all mistakes are captured

    FEEDBACK

    TEAM MEMBERS

    student Heng Yee Ying Beverly Engineering Product Development
    student Lim Yi Lin Engineering Product Development
    student Fong Jing Jie Edward Engineering Product Development
    student Edmund Ng Choon Eng Engineering Product Development
    student Tan Xiang Yun, Jason Engineering Systems and Design
    student Pang Bang Yong Information Systems Technology and Design

    © <script>alert(1)</script>

    border border