Automated Driving


CHALLENGE

According to government statistics, there are nearly 40,000 fatalities per year in the U.S. as a result of motor vehicle traffic crashes 1(NHTSA, Traffic Safety Facts 2015, published in 2017), and 1.25million annually worldwide. 2(World Health Organization, GlobalStatus Report on Road Safety, 2015). The global statistic equates to three long-range wide-body commercial jets falling out of the sky every day. Furthermore, according to the U.S. Census Bureau, by 2030, elderly adults are projected to outnumber children for the first time in the U.S. history3(U.S. Census Bureau, 2017 National Population Projects ) There are similar projections in Japan and in other parts of the world, creating an even greater need for accessible transportation for those who can not drive themselves.

TRI is showing what is possible when the limits to mobility are challenged, acting on Toyota President Akio Toyoda's call to "Start Your Impossible."TRI researchers are working to create a future where everyone has the freedom to safely move - whether that's across the country, across town or inside the home. Our insights, innovations, and collaborations strive to improve the quality of human life and shape Toyota's future products.

TRI's APPROACH

As TRI pursues Toyota's vision of "mobility for all," its work is guided by a commitment to safety in how we research, develop, and validate the performance of vehicle technology and in how we aim to benefit society.

Automated driving technology has the potential to save lives, expand access to mobility, and make driving more efficient, fun, and convenient.

TRI has a two-prong approach to vehicle autonomy, designed to give consumers a choice, which is distinct from other companies working on automated vehicles.

The first prong is Toyota Guardian. TRI's vision to create a vehicle incapable of causing a crash. Guardian's function is to actively protect vehicle occupants while the human is driving - both from errors by the driver and from external factors on the road such as vehicles, obstructions, or traffic conflicts. The higher the Guardian capability, the greater the number and types of crashes it can help protect against. For example, at the modest capability level of the current Toyota Safely Sense active safety systems, features like Lane Departure Alert (LDA) and Automatic Emergency Braking (AEB) can help prevent or mitigate some crashes. At the highest level, Guardian capability would help ensure a vehicle driven by a human being would never cause a crash, regardless of any error made by the human driver, and steer a vehicle to avoid or mitigate many crashes that would otherwise be caused by other vehicles or external factors.

The second prong is Toyota Chauffeur. The Chauffeur approach is full vehicle autonomy where there is no human driver. The vehicle operates itself and all occupants are passengers. Whereas Guardian engages control only when occupants are passengers. Whereas Guardian engages control only when needed, Chauffeur is always engaged and in control. Toyota believes that full autonomy in all driving conditions and environments (often referred to as SAE Level 5 ) is many years away. However, full autonomous operations within a prescribed set of restrictions including weather, time of day and level of traffic (often referred to as Level 4) will be available much sooner.

TRI is leveraging the power of academic collaborations to support research done in-house by its world-class research teams. TRI's three offices are strategically located near centers of innovation --MIT, Stanford, and the University of Michigan -- where it has invested millions of dollars in projects to advance technology.

HOW IT WORKS

Although Chauffeur and Guardian capabilities reflect distinct concepts for automated driving, the operation of each uses the same foundational technology. The software required for Guardian capability serves as the backbone for Chauffeur capability.

There are three elements to automated driving -- Perception, Prediction, and planning. Perception is what the automated vehicle "sees" through its technology suite of cameras and sensors. Prediction is the ability to anticipate what the surrounding traffic (other cars, pedestrians, cyclists, animals, etc.) is going to do. Planning is the decision-making process to determine what action the automated vehicles should take in response to the current environment. Of the three, the prediction component is perhaps the most challenging. Humans are very good at predicting human behavior n the road. Machines will need to be able to predict and anticipate human behavior much better. There's a tremendous amount of work ahead as an industry to develop AI for handling the infinite number of scenarios that will enable fully autonomous driving. Automated driving in most cases is actually pretty easy to accomplish, but the hard stuff is hard, especially managing the situational "corner cases."

Both Guardian and Chauffeur modes are created under Toyota's philosophy of the "Mobility Teammate Concept. " In other words, Toyota believes that people and vehicles can work together in the service of safe, convenient and efficient mobility.

TRI conducts closed course testing at GoMentum Station in Walnut Creek, Calif., and the University of Michigan's Mcity and the American Center for Mobility (ACM), both in Michigan. TRI also tests on private, gated land in Massachusetts. Additionally, TRI is constructing its own closed course facility in Michigan and will begin testing there in October in 2018. The test facility will feature the simulation of congested urban environments, a jenite pad to simulate sick surfaces and a four-lane divided highway with on/off ramps from the oval perimeter track to allow for acceleration to high speeds.

TRI uses simulation extensively to augment physical testing to account for the wide variety of situations automated vehicles may encounter in the real world. The simulation also generates potentially infinite amounts of data. Development of automated vehicle technology will require testing on the order of a trillion miles. Use of advanced simulation can help generate the data required to get there much faster.

Toyota has issued a comprehensive overview of its work on automated driving which includes a philosophical approach to the technology and a summary of ongoing research programs and near-term product plans. The white paper can be read to www.automatedtoyota.com.
 

WHAT'S AHEAD

TRI is advising Toyota in the development of "Highway Teammate" technologies, which enable driver-supervised automated driving on highways and are expected to be available on vehicles as early as 2020. The system will evaluate traffic conditions, make decisions, and take actions during highway driving. Potential capabilities include merging onto or exiting highways, maintaining or changing lanes, and maintaining distances between vehicles. Later in the 2020s. Toyota will introduce "Urban Teammate," which is expected to have Guardian and Chauffeur capabilities based upon TRI research and development. The Urban Teammate system is designed to operate in accordance with local traffic regulations for surface streets, including navigating intersections and obeying traffic lights in addition to detecting pedestrians, bicycles, and vehicles.

Further, TRI will help Toyota showcase automated driving technology at the 2020 Olympics in Tokyo, Japan.