Ola testing ‘intelligent’ driving tech

While Uber is working on building its own driverless cars and even autonomous helicopters, homegrown rival Ola is looking at more practical applications of driverless technologies.

The Bengaluru-based company has quietly started testing how it can prevent an accident by alerting the driver if he/she seems a bit drowsy, or if the car seems likely to hit an object ahead. The technology will also be able to alert a central command in case of a collision to send assistance immediately while informing the passenger’s friends and family about the incident.

“We are bringing different signals together — from the car, from the Ola ride history, booking experience of the day and location in the city, such as accident-prone area — and using these data to prevent accidents,” Ankit Jain, VP and Head of Ola Play, told BusinessLine .

The ride share company is trying to improve passenger safety in two steps. In the first phase, it is looking at generating alerts in case of an accident, to call for assistance.

In the second phase, where the company intends to use autonomous driving tech such as cameras, radars and telematics, the aim is to detect collision before it takes place and take action to prevent it.

“We will be able to detect if a driver is drowsy, either through a camera or by figuring out how the person is driving. If the driving pattern is unusual, we can take action, depending on our confidence on how drowsy the person is. For example, it is dangerous if the driver has been working for 13-14 hours late into the night. The situation will be handled differently if it were in the morning,” Jain said.

For this, Ola will capture data from the car and relay it to a command centre in real time. This info will be combined with the driver’s daily drive data, such as his/her work duration and if he/she’s driving in an accident-prone area. Added to that, camera and radar sensors placed in the car will be able to detect collision.

However, these sensors will not be as accurate as the ones used in driverless cars, where the margin of error is zero. However, 95 per cent accuracy can go a long way in saving lives, says the firm. Plus, these sensors are far cheaper.

“We are focussed on how not to use expensive sensors but to use embedded devices with additional cameras, which can be fairly cheap,” Jain said. Early experiments suggest that the method is effective, he added.

These technologies are already being tested by Ola on its Ola Play fleet, which is fitted with several sensors that collect key data. The company will soon add cameras to this fleet to enable collision prevention.

So who will bear the additional cost for safety? Ola is trying to distribute it between itself, carmakers and insurance firms.

If tests go as planned, we might see some of the safety features in Ola Play cabs in two months. within Meanwhile, the company is in talks with carmakers to get access to more data from the car.

COMMENT NOW