Maruti Suzuki In Progress With Microsoft AI Based Technology
Harnessing Automobile for Safety (HAMS) technology has been developed by Maruti Suzuki India to process data through automated assessment of driving skills. Such technology improves accuracy wherein video analytics are assessed to determine the driving efficiency by a smartphone based solution. Artificial Intelligence has transformed the whole world with its integration through machine learning ensuring real-time processing of data enhancing accuracy compared to the manual indulgence in data management. Cloud Technologies combined with Cognitive Intelligence has upscaled Robotic Process Automation (RPA) enabling ease of performing complex tasks more accurately and efficiently.
The technology, HAMS has been deployed at the Automated Driving Test Centre (ADTC), Dehradun, in association with the Transport Department, Govt. of Uttarakhand - as per the Microsoft Research India (MSI) - after being tested jointly in Maruti Suzuki promoted Institute of Driving and Traffic Research (IDTR). MSI Executive Advisor, Ajay Kumar Tomer stated that,HAMS, developed in partnership with Microsoft Research India, is an important tool to modernise the driving license test in our country. Starting with ADTC Dehradun, this technology will be introduced in more centres across several states.Such technology has not only decreased the total time taken in assessment of the driving test to around 10 minutes compared to manually generated reports, however has also reduced the pass rate to only 54% compared to previous 90% - due to accuracy and transparency involved in the whole process.
Harnessing Automobile for Safety (HAMS)
The smart sensing technology used, employs the smartphone cameras in analysing situations pertaining to the effective detection of driver's emotional changes causing any distractions through the front camera and monitoring road activities simultaneously using the rear camera. Even the braking event is detected using the mobile's accelerometer, addressing to the most complex challenges through advanced integration of machine learning models. As stated by the Microsoft research project report,
As part of the project, we have also explored several use cases for HAMS. One of the earliest we prototyped was a fleet management dashboard, which allowed a supervisor to view safety-related incidents of interest offline.Such technology does not restricts its usage in driving test assessment, however has the expertise to detect the driving lane wherein the driver is informed with the suitable speed to be maintained and length of the lane, unlocking wider horizons compared to a Map. This auto detection technology known as 'DeepLane' has an accuracy of over 90% in determining the lane's position. Features employed in the technology providing an edge to innovation are - • FarSight: Smartphone Based Vehicle Ranging • DeepLane: Computer Vision based Lane Detection • FullStop: Tracking Unsafe Stopping Behaviour of Buses • AutoRate: Automatically Rating Driver Attentiveness • InSight: Driver State Monitoring in Low-Light Conditions • ALT: Automatic Driver License Testing • Application: Fleet Monitoring. HAMS has the potential to minimise accidents arising due to the casualness of driving leading to a more advanced monitoring system that can be used by Governments ensuring road safety measures to be followed strictly. Apart from the driver's fault, features as - FarSight/ Fleet Monitoring detects any potential misappropriate behaviours of other vehicles around to alert the driver with safety protocols to be self adopted.