Not wishing to miss the boat of putting driverless cars on roads by 2025, Indian companies, startups and IITs have joined the race of building the next autonomous vehicles technology in India. Leading the Indian charge is Tata Elxsi, Tata’s leading design and technology services arm for product engineering across automotive, communications and broadcast. Tata had earlier gained significant ground by debuting its Autonomous Valet Parking prototype that demonstrated the fusion of sensor and vision based ADAS technology.
Now, reportedly, Tata Elxsi has made a big leap by testing driverless cars on Bangalore’s roads. According to news reports, Tata in collaboration with its rich ecosystem of component suppliers is leveraging AI to develop driverless cars and has stolen a march over its Indian competitors (Mahindra, notably) by being the first company to test driverless vehicles on Indian roads.
Driverless cars – the future of mobility
According to Chetan Maini, founder of Reva, India’s first electric car and director at Maini Group, “The DNA of the auto industry is changing. Vehicles are getting smarter, more connected and possibly way more virtual. We are trying to rethink where mobility is going to go”
Maini believes the challenge in India is more unique. “Can we have that’s driverless and work for India, in Indian conditions, imagine the solution for tens of millions differently able people today in the country who can use driverless mobility to get them around and can the solution be far more affordable?” he said.
Maini played out a simple scenario that could become a reality in near future: Imagine pressing a button on your mobile phone, your car comes up to you, you don’t have to think about parking, you get in it, and suddenly someone steps in front of you and your car’s braking system takes over and stops before you can even react. You drive to your destination, get out of it, press another button, and it goes and finds its parking spot, so that’s the innovative solution we are looking for as part of Mahindra’s Rise challenge.
Mahindra’s driverless car challenge sees 13 Indian startups including IITs rise to the occasion
Mahindra is betting big on the future of mobility and believes autonomous cars is the next evolution for the automotive industry. The paradigm shift will be seen in increased environmental friendliness, overcoming parking problems, decreased mortality rate and a differently-abled friendly transportation.
Self-driving cars are outfitted with radar, GPS and computer vision and most importantly LIDAR or the light radar, the sensing technology that helps driverless vehicles in precise localization. In fact, LIDAR sensors could significantly bring down the cost of autonomous vehicles and make it affordable for average buyers.
Mahindra’s Rise prize Driverless Car challenge with a purse of $1 million has seen Pune-based Aeron Systems Pvt Ltd, maker of IoT devices and inertial sensors and systems, AGV IIT Kharagpur, Auro Robotics, drawing members from IIT Kharagpur and CMU Robotics Institute, IIT Bombay, d-LIVE the official team from IIT Delhi, Team Manas from Manipal Institute of Technology, Pragyaan backed by DRDO make the final cut to top 13 for the prototyping phase. Amongst the startups, Delhi-based AI start-up Cube26, known for developing customized OS platform for OEMs and OmnipresentRobot, one of leading robotics and drone makers in India that leverages computer vision, machine learning and virtual reality are now taking a crack at driverless vehicle technology.
IoTIndiaMag lists down some notable projects underway
OmnipresentRobot: According to news reports, Akash Sinha led startup that also has CMU’s Dr Raj Reddy, eminent computer scientist and robotics experts is well positioned to ace the challenge given Sinha’s participation in three 3 DARPA Robotics Grand challenges. Taking cues from DARPA, the startup’s modus operandi is using a combination of ID Laser sensors, stereo cameras and combining their data through sensor fusion to generate 3D maps. There are also plans to deploy monocular camera for detecting traffic lights and decoding road signs. Deep learning techniques will be leveraged to enhance vehicle performance and train it further to understand barricades, dividers and even Indian hand gestures used during driving. Finally, there will be an app to follow instructions such as parking, go to address location. What the startup is also striving hard is achieving precision localization through ultra wide band tags along with GPS/INS.
Auro Robotics: Auro robotics backed by IIT-Kharagpur alumni have already made significant headway with their autonomous shuttle currently playing in the University of Santa Clara with multiple campus trials ongoing. The driverless electric shuttles address the “last-mile” challenge in public transit systems, says Nalin Gupta, CEO of Auro Robotics. Their electric shuttle, relies heavily on LIDAR, sensor technology that uses laser light for localization and spotting anomalies.
SeDriCa 1.0 from IIT Bombay: SeDriCa is the famed Autonomous Ground Vehicle, cranked out by IIT Bombay’s Innovation Cell that nailed the 4th spot globally in the Intelligent Ground Vehicle Competition (IGVC) 2016, besting teams from CMU, Virginia Tech. and 30 other universities. Now the team is applying their learnings to Mahindra E20. The task at hand – autonomously maneuvering Mahindra’s electric car on Indian roads. From pedestrian detection to positioning system, the team has a similar approach, combining sensor data from leveraging AI to perfect autonomous vehicles. According to their home page, the team’s looking at GPS/INS, LiDAR and stereo cameras to gather information about the immediate environment at the right range. The 12-member team is led by Shubham Jain, a 4th year Mechanical Engineering student who also wears the hat of Project Manager and Deputy Team Leader.
Team Manas from Manipal University: Combining AI, automation sensing and mechanical engineering, a team of 45 undergraduates from Manipal University are experimenting with lane detection, maze traversal, for navigation and parking and much more. From performing real world 2D LIDAR obstacle detection to carrying out dynamic obstacle avoidance to sharpening localization, Project Manas team is at hard at work to best their competitors.
Google, Uber proclaim self-driving cars distant reality in India
Driverless cars may not be favorable for Indian roads, what with Google’s CEO Sundar Pichai and Uber’s founder and CEO Travis Kalanick deeming Indian roads, the last place for self-driving cars. In India, a lack of lane discipline coupled by traffic conditions and a large number of two-wheelers on roads present certain challenges. The traffic on Indian roads cannot be easily mapped by autonomous vehicles. The present condition of Indian roads – potholes, poorly developed roads, obstructions and traffic diversions present major challenges to autonomous vehicle technology.
Experts proclaim that driverless vehicle technology is bound to become ubiquitous in developed nations by 2025 with legacy automotive companies Tesla, Apple, Baidu, Tesla, Ford, Japan’s Nissan ,Honda, Mitsubishi and Toyota and Germany’s Audi and BMW, putting their might behind self-driving cars.
From cars to pods – ushering in a revolution
Interestingly, California, the hotbed of driverless technology where it has become commonplace to spot autonomous vehicles plying on the road, could very well become the first state to allow driverless vehicles on empty roads. And the future of cars will be a pod, possibly with a camera mounted on top to collect all the data.
America’s biggest car market is proposing new regulations wherein road ready vehicles could be tested without drivers could start by the end of 2017. If the regulations are passed, a limited number of cars will be in the market as early as 2018. Google’s rechristened self-driving unit, Waymo is an everyday sight in Silicon Valley and has had successful real world tests. Interestingly, Google and Uber are battling out over driverless technology, with the former alleging Uber stole its technology to get ahead in the race.