Nissan's Self-Driving System Will Adapt to Local Driving Cultures and Societies
【Summary】This is crucial during the initial release of self-driving cars, since they will share the road with SAE L1 non-autonomous vehicles, or cars driven by humans with driving skills heavily influenced by local communities.
There are many ways to drive on public roads. In the US, most drivers carefully follow road regulations, abide closely to speed limits and avoid using the car horn. But in other parts of the world, like some countries in the Middle East and Asia, many people don't use turning signals diligently and pedestrians cross almost anywhere they want to on congested streets.
For driverless technology, this could be a huge problem. When refining autonomous driving systems, developers must take local driving practices into consideration. This is crucial during the initial release of self-driving cars, since they will share the road with SAE L1 non-autonomous vehicles, or cars driven by humans with driving skills heavily influenced by local communities.
Nissan, a pioneer in driverless technology, wants to address such issues using insights from sociologists and anthropologists. The group plans to use the information to hone their system on an algorithmic level, which will cause the vehicles to become more "sensitive" to their surroundings.
Adjusting to Social Driving Norms
Perhaps the most difficult aspect of adjusting to social driving norms is that some local practices overlook road regulations and guidelines. For example, highway driving on US roads entail going roughly 10 mph faster than the average speed limit in the appropriate lane. As Google pointed out previously, going slower than cars that are breaking the speed limit on public roads could leave an autonomous vehicle more prone to collisions. With this in mind, a self-driving system that fails to relate to how other cars are behaving may compromise the safety of passengers. This is a risk that Nissan hopes to reduce by studying traffic patterns, pedestrian movements and activities of drivers.
"The way people drive in Sao Paolo differs from the way people drive in Silicon Valley, California. What is socially acceptable in Sao Paolo is not socially acceptable in Silicon Valley. Autonomous cars will need to take this into account," said Maarten Sierhuis, director at Nissan's Research Center in Silicon Valley.
Nissan is enlisting help from social experts due to the prohibition of driverless cars on open roads in countries where the company is firmly established, like Japan. As a result, the automaker is conducting closed pilot programs in various locations around the country and its own manufacturing facility (Nissan Oppama plant) in Yokohama, near Tokyo. According to Japanese law, it is legal to utilize self-driving vessels in private establishments. Inside the plant, the company deployed a fleet of modified autonomous Nissan Leaf vehicles to tow cars around the area.
The vessels are capable of stopping for other cars in the facility. However, they are still far from the final stages of development. During a demonstration, one of the vehicles did not engage the road due to a glitch. To mitigate such errors effectively, Kazuhiro Doi, Nissan vice president, confirmed that mechanical driving components are available inside the cars for developers to step in and make minor adjustments.
"There are patterns that we can perhaps prime into the cars so the systems will be able to interpret the environment. With a social lens, we can do this without relying on miles and decades of technological sensing," explained Melissa Cefkin, principal scientist and design anthropologist at Nissan Research Center.
Michael Cheng is a legal editor and technical writer with publications for Blackberry ISHN Magazine Houzz and Payment Week. He specializes in technology business and digesting hard data. Outside of work Michael likes to train for marathons spend time with his daughter and explore new places.
Zomato Tests Drones for Food Deliveries in India
Alibaba Brings Tmall Genie to Audi, Honda and Renault Vehicles
Driverless Sensor Startup Sense Photonics Raises $26 Million in Series A Funding
Startup Alaka'i Unveils Hydrogen-powered Flying Car
BMW and Jaguar Land Rover Collaborate on Electric Vehicles
Waymo’s Autonomous Trucks Enter New Testing Phase in Phoenix
Didi Chuxing Partners with China’s State Grid, Explores EV Services
GM and Bechtel to Build Massive EV Charging Network
- Study Reveals Consumer Views on Driverless Car Makers, Tech and Trust
- Daimler Acquires Stake in Sila Technologies to Develop Next-Gen EV Batteries
- Autonomous Delivery Startup Gatik Emerges From Stealth, Announces Partnership with Walmart
- Alibaba Brings Tmall Genie to Audi, Honda and Renault Vehicles
- The U.S. is About to Face a Battery Shortage in Race to Ramp up EV Production
- Lyft Valued $24.3 Billion Ahead of Tomorrow’s IPO, Becoming the First Ride-Hailing Company to Go Public
- Uber Looking to Sell a Minority Stake in its Autonomous Vehicle Unit
- Waymo to Outfit its Driverless Vehicles in a Former Detroit Axle Factory
- China to Scale Back EV Subsidies to Push for Better Cars
- Apple Reportedly Making a Bid to Purchase Drive.ai