Industrial Consultancy & Sponsored Research (IC&SR) , IIT Madras
Technology Category/ Market
Category –Automotive
Applications – Transport systems, Automation, Automobiles
Industry –Automotive/ Transportation Systems
Market -The global intelligent transportation system market is projected to grow from $22.91 billion in 2021 to $42.80 billion in 2028, at a CAGR of 9.34%
Image Gallery
Problem Statement
- Manual toll collection has limitations such as human error, missed record of vehicle details, slow process, customer irritations, misbehaviour by the driver or the toll attendant, theft, other disturbances
- Tag based system on the other hand needs participation and associated issues
- These can be resolved with a toll system that can accurately classify the vehicles with an automated system that does not need participation by keeping a tag in the vehicle
Technology
A method and system for robust classification of vehicles under heterogeneous traffic condition, comprises of:
- Hardware-executable components for execution
- Inductive classification loop sensors
- A smart inductive loop
- Graphical User Interface
- Microprocessor
- Display Module
- Smart inductive loop is positioned between the classification loop array and the payment summit, it employs time series techniques and machine learning algorithms to process the collected data
- The inductive classification system is designed to classify heterogeneous traffic flow, with one loop per lane of a modified shape with inner and outer loops having specific dimensions.
- The said microprocessor enables signal de-noising using DWT technique; signal segmentation with moving standard deviation technique; feature extraction with DWT and vehicle classification using SVM classifier
- Display Module displays information related to accounts associated with the operation and classification of vehicles
Key Features/ Value Proposition
Technical Perspective:
- A robust mechanism in classifying vehicles under heterogeneous traffic conditions, which the regular loop designs and algorithmic methods cannot classify
- Signal De-noising using DWT (Discrete Wavelet Transform) technique helps to remove noise from the collected signal data
- Vehicle Classification using SVM (Support Vector Machine) classifier that classifies vehicles based on the extracted features.
User Perspective:
- Enhanced, efficient and improved inductive loop for vehicle classification
- Accurate in obtaining the vehicle signatures from the de-noised data
Questions about this Technology?
Contact for Licensing
sm-marketing@imail.iitm.ac.in
ipoffice2@iitm.ac.in
Research Lab
Prof. LELITHA DEVI V
Department of Civil Engineering
Intellectual Property
- IITM IDF Ref. 1802
- IN201941013771
Technology Readiness Level
TRL- 5
Technology validated in relevant environment