Global AI Conclave: Bengaluru traffic management to fully shift to AI-based systems, says traffic police chief

Anucheth was speaking at the CNBC TV18 - Moneycontrol Global AI Conclave, moderated by Moneycontrol Deputy Executive Editor Chandra R Srikanth, in Bengaluru on November 22.

He said that 96 percent of traffic violation cases are being booked through contactless enforcement. “Under the Intelligent Traffic Management System (ITMS) project, we have deployed AI-based cameras at 50 junctions to detect seven types of violations, including over-speeding, mobile phone use while driving, not wearing seat belts, not wearing helmets, triple riding on two-wheelers, red light violations, and stop-line violations,” Anucheth explained. “Helmet violations are the most common, accounting for 70 percent of the cases.”

Also, read: Bengaluru now books 96% traffic violations through AI-powered cameras

He said that AI cameras are significantly more efficient than human officers, with one AI-based camera able to detect 3,000 to 4,000 violations at a junction in the time it would take 10 constables working in four shifts to cover.

“We are now expanding AI-based enforcement system, incorporating 9,000 Safe City cameras. We are also building an AI platform that will detect 19 types of violations, up from existing seven from January 1. We will also expand it from 250 junctions to 1,000 junctions within a year,” he said.

Also, read: Bengaluru Traffic Police to use AI for 13 different violations, up from seven

Anucheth said that Bengaluru has already implemented AI-based traffic signals.“At present, we have two types of AI-based signalling systems in place: C-DAC (Centre for Development of Advanced Computing)’s indigenously developed CoSiCoSt ATCS application, which is installed at 72 junctions and will reach 165 junctions by the end of February. We also have Japanese MODERATO technology at 28 junctions. In a couple of months, we will have roughly 200 junctions covered across the city,” he said.

He said AI-based signals adjust timings based on traffic flow, which has helped reduce travel time by 17-22 percent, as shown by GPS tracking through a third-party audit.

Also, read: AI-powered signals reduce travel time by 20-33% across three road sections: Bengaluru Traffic Police

Bengaluru's vehicle population has surged from 5.6 million in 2013 to 12.4 million in 2023.

On AI avatars being used by the Bengaluru police on social media, Anucheth said: "We are using AI to create videos that replicate my voice and appearance. This will help us to engage with the public in a more effective, appealing way than traditional press releases."

Also, read: AI Cops! Bengaluru police pilots AI-generated avatars for social media outreach

On the misuse of AI avatars, he said, “There are many checks and balances in place, with approvals required for the final cut. We ensure that only after our approval is the content used. As with any new technology, misuse is possible, but we are guarding against it by implementing strong safeguards.”

ASTraM app launch

Bengaluru traffic police will soon launch ASTraM (Actionable Intelligence for Sustainable Traffic Management) mobile application, which will provide real-time traffic and weather updates.

“This app will help citizens plan their routes for 6 to 24 hours in advance. It will also include live updates on road conditions, traffic congestion, accidents, and road closures.” he said.

Also, read: Bengaluru traffic police launches AI-powered ASTraM; to provide congestion alerts

Anucheth said that AI is not only improving traffic flow but also helping with congestion prediction and monsoon management. “It will predict and monitor congestion in real-time, leveraging data from CCTV cameras, travel aggregators, and even weather services. We can now predict where waterlogging might occur, allowing us to issue travel advisories,” he said.

Trevor Noah impact

Bengaluru traffic police faced flak after comedian Trevor Noah cancelled his September 27, 2023 show in Bengaluru, citing "technical issues," with many attendees blaming traffic congestion on Outer Ring Road as one of the contributing factors.

“It was a perfect storm that day—heavy rainfall, a long weekend, and six vehicles breaking down on the Outer Ring Road, causing hours of delays,” said Anucheth. “This gridlock highlighted the importance of monsoon management in Bengaluru’s traffic strategy, leading to the development of ASTraM platform."

Also, read: 5 reasons why Bengaluru's ORR faced unusual traffic congestion on September 27

ASTraM uses real-time data from multiple sources, including CCTV cameras, travel aggregators, and weather data from the IMD, to predict congestion and waterlogging.

"We now track congestion at a granular level, making real-time decisions, such as staggering bus timings during peak hours." He added that while AI can’t solve all problems, "It’s been crucial for improving traffic management outside of monsoon-related issues".

He said that Wednesdays are the most congested day of the week. “Busiest days in Bengaluru are Tuesdays, Wednesdays, and Thursdays, but congestion is significantly higher on Wednesdays between 6:30pm and 7:30pm.”

Anucheth said that AI and technology will play a key role in managing Bengaluru’s traffic but stressed the need for public transport adoption, stating, “No amount of AI or infrastructure will solve congestion unless we integrate public transport into our daily lives.”

Also, read: Why Wednesdays and Thursdays are the most congested in Bengaluru