While we may think of ourselves as expert drivers, there will always be factors - like inclimate weather - that we simply can’t control. To help ensure our safety, Toyota has supported researchers around the world in exploring how to augment the skills of human drivers with artificial intelligence (AI) that can help us to be more aware of our surroundings and prevent some accidents from occurring.
Enter “Guardian,” the Toyota Research Institute (TRI)’s cutting-edge surveillance technology that could assist drivers by helping to monitor the environment around their vehicle. When a potential hazard emerges—such as an unexpected jaywalker or an upcoming patch of ice—Guardian is able to alert the driver in advance and can even step in to protect them if it determines they’re unable to respond in time.
The innovative Guardian system uses LiDAR lasers to detect imminent dangers. It works much like radar, but instead of radio waves, it deploys pulses of invisible infrared light. These pulses are emitted from the vehicle and bounce off of nearby objects at a rate of a million per second. Once they reflect back into the vehicle’s sensors, they’re then combined into what’s called a “point cloud,” or a 3D map of the surrounding environment. The maps’ staggering level of detail can identify objects, predict how they will behave, and then help the driver and car respond accordingly—all in a matter of milliseconds.
“Imagine you're going through an intersection and someone is going to run a red light and T-bone you. We imagine our Guardian technology to be able to understand and predict that, and to be able to accelerate you out of the way,” Ryan Eustice, vice president of autonomous driving and head of TRI's office in Ann Arbor told Autoweek.
Guardian doesn’t just have its laser-guided eyes on the road. It can also monitor the driver’s behavior inside the vehicle. An infrared sensor mounted on the steering column can determine if the driver is showing signs of drowsiness or distraction, and it can even recognize droopy eyes hidden behind a pair of sunglasses.
"If the driver has a 99 percent chance of detecting hazards and the [Guardian] automation system also has a 99 percent chance of detecting hazards, that gives the combination of the driver and system a 99.99 percent chance," Steven Shladover, researcher at the University of California, Berkeley’s PATH program told the MIT Technology Review. “This is much simpler and easier than designing a fully automated system that could reach that 99.99 percent level by itself."
Guardian is one half of a pair of revolutionary AI safety technology pathways currently being developed by Toyota. While the Guardian technology would serve as a trusted back-up in case of emergency, its counterpart, Toyota’s “Chauffeur” technology stream, would be a full-service AI driving system that could navigate roads without a person supporting it at the wheel. In the future, Chauffeur technology would enable a truly autonomous vehicle that would navigate diverse terrain with the skill of its human counterparts.
The Guardian and Chauffeur systems use the same sensors and camera hardware so, theoretically, vehicles equipped with Toyota’s autonomous tech could incorporate Guardian, then “graduate” to Chauffeur when that more complex system is eventually perfected. Toyota’s choice to explore parallel technology tracks and a graduated process, rather than focus its research energy exclusively on racing to achieve a fully automated vehicle, could have significant benefits for our safety and wellbeing in the meantime.
Toyota has announced that it will demonstrate TRI’s ambitious vision - both the Chauffeur and Guardian technologies - at the 2020 Olympics in Tokyo.
