Mountain View, California-based Drive.ai is a startup created by former lab mates from Stanford University’s Artificial Intelligence Lab. Originally founded in 2015 by Carol Reiley and Fred Rosenzweig, Drive.ai raised $12 million in Series A funding earlier this year to develop deep learning algorithms to control the operation of autonomous vehicles. In the process of developing driving algorithms, they may also back into general artificial intelligence (AI).
Why There Is a Need
Building on experience gained from the DARPA Grand Challenge, Google and other self-driving pioneers programmed the first self-driving car to rely primarily on light detection and ranging (LIDAR), which is a remote sensing method that uses pulses of laser light to measure distances, and detailed mapping. Although this has worked pretty well, the current technology is expensive. Drive.ai’s approach is to use low-cost cameras and deep learning to program cars to drive themselves.
Benefits of Autonomous Vehicles
Making autonomous vehicles easier to manufacture with less expensive parts will make them more affordable. There is already a demand for self-driving cars from 45 million elderly people, 21 million people that are vision impaired, 18 million children between the ages of 12 and 16, and 12 million alcoholics, so lowering the cost will speed adoption of the technology with all the benefits of reduced traffic, parking, road construction, and fuel consumption. Drive.ai also claims it can alleviate jerky driving.
Challenges of Artificial Intelligence
Cameras are inexpensive, but it is hard to translate images into detected objects. In response to this challenge, Drive.ai is developing a multitasking neural net. Deep learning is a general-purpose pattern recognition system that works with any data where there is a statistical correlation between the input and the output. Pattern recognition is an important part of natural intelligence, but only a part.
Strong AI (which is now being rebranded as general AI) attempts to simulate general human thought processes by using a computerized model of concepts to organize knowledge and then act on them. The goal is to perform any intellectual task that a human being can perform. Instead of being programmable in the traditional sense, strong or general AI seeks to make sense of the world by relying on the perceptual experience of changes in the physical world, buttressed by rules and the discipline of logic. Advances in neural science and the recent success of convolutional neural networks (CNNs) suggest that human intelligence is an amalgamation of different forms of intelligence, although there is evidence that all biological intelligence uses similar structures. Tractica’s Artificial Intelligence for Enterprise Applications report also defines strong AI as integrating different modes of AI. Multitasking neural nets seem to be a step in the direction of general AI.
Bright Future for Drive.ai and Others
Companies like Drive.ai provide support to Tractica’s position that the automotive industry is poised to leverage AI extensively in the future. Tractica’s Artificial Intelligence for Enterprise Applications report forecasts that annual spending in this sector alone will grow to nearly $1 billion by 2025, which is a bright future for companies like Drive.ai that are entering the market.