Optional extra: The ‘cow-avoidance’ driver assistance system
Two Indian students propose an obstacle avoidance system for a very culturally-specific problem.
It’s not exactly something that those of us in the Western World have to worry too much about. We may have been stuck behind a herd of cows on a quiet rural road once or twice – but encountering these bovine beasts in the middle of busy roads is not a daily occurrence.
In India, however, it’s a very real problem. Why? Because 80% of the country’s 1.2 billion-strong population are Hindu – a religion that holds cattle in such reverence that they are not interfered with as they mix with rickshaws and cars alike. Vehicle-animal collisions are therefore frequent – and are said to be a contributing factor to the high number of deaths by traffic accident in the country: estimated at over 200,000 in 2013, according to the World Health Organization. Indeed, such poor road safety records is exactly what automated driving is trying to eradicate.
India hasn’t featured heavily in automated driving news since it boasted a driverless shuttle bus at the Indian Auto Expo last year, but perhaps students Sachin Sharma and Dharmesh Shah from the Department of Electronics & Communication, Gujarat Technological University are on to something. They recently submitted a paper to the International Journal of Vehicle Autonomous Systems outlining a cow-specific object avoidance system.
Inspired by advanced driver assistance systems, the presumably retro-fitting setup uses a dashboard camera as well as an algorithm that can assess if an object is in fact a cow and if its movements represent a risk to the vehicle. In the paper abstract, the pair said:
“We designed a system based on histograms of oriented gradients and cascade classifiers for cow detection. The Indian cow has been the biggest obstacle compared to other animals on Indian roads. The distance between a cow and the vehicle is calculated prompting an alert signal to notify the driver for applying brakes. The method is implemented in OpenCV software and tested on various video clips. The proposed system has achieved an accuracy of 80 percent in terms of cow detection.”
Safety is not the only challenge India faces when it comes to their roads. The country, which is home to 13 of the world’s 20 most polluted cities, could also benefit from the reduced emissions that automated driving would bring about. And what’s more, it’s people seem to be open to the idea according to a 2015 survey where 85% of the Indian respondents had an overwhelmingly positive attitude to the technology.
Read the full International Business Times article here.