When it comes to ground transportation, most of the research and development on GNSS is aimed at developing driver assistance systems and ultimately driverless cars and trucks. To that end, GNSS receivers are integrated with inertial navigation systems, radar, lidar, computer vision, and ultrasound.
Leveraging decades of robotics experience and knowledge of control algorithms, AutonomouStuff, part of Hexagon’s Autonomy & Positioning division, has developed a software stack for autonomous vehicles based on the Apollo open source software stack. .
Kevin Fay, product manager for Hexagon’s platform and vehicle software business, said: The software stack can be customized across platforms to meet your equipment needs.
Most recently, in a joint project with the University of Iowa’s National Advanced Driving Simulator, AutonomouStuff worked with the Automated Driving Systems for Rural America project to equip a Ford Transit 350HD shuttle for autonomous driving. First, we created a drive-by-wire system that allows electronic control of the vehicle, then installed position, navigation, and recognition sensors. As a result, the platform is ready to become autonomous as soon as the software stack is integrated.
Rural roads, which have a wider range of speeds than city roads, can be blocked by wildlife and heavy machinery. And with surfaces ranging from asphalt to gravel, it provides a particularly challenging testing environment for autonomous driving software.
“Vehicles in Iowa have done a fair amount of self-driving on a combination of urban and rural roads where traditional sensing doesn’t work,” says Fay. “It excels in areas such as gravel roads with limited or no lane markings or narrower than normal. can now be properly navigated.”
Rural roads generally don’t have the GNSS multipath challenges that urban canyons pose, but they also have fewer navigational landmarks. Another challenge is bad weather. Faye noted that country roads may be plowed during snowstorms. “If you’re always in the right lane of the road, you can get out of the ruts in the road and have a hard time getting through.” The vehicle learns to navigate properly in those situations. need to do it.
The University of Iowa Ford Transit Shuttle is a limited deployment and collects data primarily for research purposes. All the while there is a safety driver, but providing residents with a real ride. “They’re always paying attention, but their hands are next to the steering wheel,” he says. “There will be times when they will have to take over.”
Other universities and companies are using the platform to advance their autonomy programs. Most of them are urban driving on complex routes with heavy traffic, with a total of 12 vans in use across the country.
Hexagon equips its vehicles with a variety of sensors, including front-mounted adaptive radar, roof-mounted Velodyne lidar, roof-mounted NovAtel GNSS receivers, and interior-mounted cameras. “What software we offer depends a lot on the customer and the software they’re deploying,” he says. “We offer our customers a complete package that can be used right out of the box with minimal work. It has the software, the interfaces to the vehicle and the sensors. We can provide a vehicle with , and add our own computers and software on top of it.”
Hexagon’s first Ford Transit will be delivered in 2021. The company will release his current version in the spring of 2022, and the Iowa project will run until his mid-2023. “So you can continuously grow your skill set and overall expertise.”