Solutions powered by artificial intelligence (AI) may soon make Indian roads safer to drive. A unique AI approach that uses the predictive power of AI to identify road hazards, and a collision warning system to communicate timely alerts to drivers, to deliver several road safety related improvements , is being implemented in the city of Nagpur with the objective of bringing about a significant reduction in the number of accidents.
A project, “Intelligent Solutions for Road Safety through Technology and Engineering” (iRASTE), has been launched to identify scenarios that can cause accidents while driving a vehicle and alert drivers about them using the Advanced Driver Assistance System (ADAS).
The project will also identify ‘grey spots’, i.e. through data analytics and mobility analysis by continuously monitoring dynamic risks across the entire road network. Gray dots are locations on the roads, left unaddressed could become black dots (locations with fatal accidents). The system also performs continuous road monitoring and designs technical fixes to correct black spots in existing roads for preventive maintenance and road infrastructure improvement.
The iRASTE project is being carried out by I-Hub Foundation, IIIT Hyderabad, a Technology Innovation Center (TIH) set up in the technology vertical sector Data Banks and Data Services supported by the Department of Science and Technology (DST) as part of its national mission on interdisciplinarity. Cyber-Physical Systems (NM-ICPS) with INAI (Applied AI Research Institute). The project consortium includes CSIR-CRRI and Nagpur Municipal Corporation, with Mahindra and Intel as industrial partners.
The Hub strives to coordinate, integrate, and amplify basic and applied research in data-driven technologies, as well as its dissemination and translation across the country. One of the main objectives is to prepare an essential resource for future use by researchers, startups and industry, mainly in the fields of smart mobility, health and smart buildings.
What makes the iRASTE project even more unique is that AI and technology are applied to create practical solutions, as a model, for Indian conditions. While the initial deployment of iRASTE is in Nagpur, the end goal is to replicate the solution in other cities. Currently, talks are underway with the Telangana government to adopt the technology in a fleet of buses that ply the highways. There are also plans to extend the reach of iRASTE to Goa and Gujarat.
I-Hub Foundation has also used techniques ranging from machine learning, computer vision, and computational sensing for several other data-driven technology solutions in the mobility industry. One such solution is the India Driving Dataset (IDD), a dataset for understanding road scenes in unstructured environments captured from Indian roads, which stands out for deviating from global assumptions of driving. well-delineated infrastructure such as lanes, limited traffic participants, low variation in an object or background appearance, and strict adherence to traffic rules.
The dataset consists of 10,000 images, finely annotated with 34 classes collected from 182 driving sequences on Indian roads obtained from a front camera attached to a car driving around Hyderabad, Bangalore and their periphery. The dataset is released in the public domain for unlimited use under public license and becomes a de facto dataset for all analysis on Indian road scenes. Currently, there are more than 5000 registered users for this dataset across the world.
Another dataset called Open World Object Detection on Road Scenes (ORDER) has also been developed using the India Driving Dataset which could be used by autonomous navigation systems in Indian driving conditions for localization and the classification of objects in a road scene.
In addition, a mobility car data platform (MCDP) has been designed with multiple sensors – cameras and LIDAR, with the necessary computation for anyone to capture or process data on the car – can help researchers and start-ups. -ups in India testing their automotive algorithms. and approaches to navigation and research on Indian roads.