Diving into the exciting, new world of autonomous systems technology, Bell’s Innovation team continues to lay the foundation for autonomous vehicle management, which we anticipate to be in high demand as the infrastructure for unmanned vehicle systems grows. At CES 2020, Bell unveiled Bell AerOS, which enables companies to manage their fleet information, observe the health of their vehicles, manage throughput of goods and products, and aggregate predictive data and maintenance information, built on the Microsoft Azure cloud platform. The team stunned the crowds at the world’s largest consumer tech convention by revealing a very possible future, involving seamless flight management whether carrying goods or people.
For 85 years, Bell has proven itself as an innovative leader in aerospace. The company has been committed to achieving the impossible, and this commitment has enabled it to bring to life some of the world’s most famous aircraft: The Huey and V-22 Osprey. To help in this mission, Bell is working with Microsoft autonomous systems to realize its vision of intelligent unmanned aircraft.
The challenge of building intelligent unmanned aircraft can be broken down into smaller problems that can be worked on individually, like take-off, flight, and landing. While the long-term objective is to have all those stages of flight completed by AI, Bell chose to start with the challenge of landing. Landing has two key components: identifying safe landing zones and completing the landing procedure. For a person, a safe landing zone seems straightforward: Look for a decent-sized flat area that isn’t obstructed. However, this problem is much harder for AI because landing zones can appear differently based on the aircraft’s altitude.
For example, from a few hundred feet up, a field of wheat may look like a safe flat landing zone. However, if the AI doesn’t realize that the top of wheat is not firm ground, it may execute a landing manoeuvre several feet too early – causing damage to the drone, its cargo, and any passengers. Bodies of water pose a similar problem. Additionally, when a large, unmanned vehicle lands, the dust-up from the rotor blades can cause vision to be obscured completely.
The Bell Intelligent Systems team identified Microsoft autonomous systems technologies that could help solve the problem: Microsoft Project Bonsai and AirSim. Along with computer vision and cloud computing from Azure, autonomous systems technologies help Bell create an AI that can practice in a simulated environment.
The project’s great success is owed to a close collaboration between Bell and Microsoft. Continuing on their exploration of intelligent systems, today Bell is working on autonomous precision landing using Project Bonsai, a Microsoft autonomous systems initiative. The system will enable Bell’s vehicle to identify safe landing zones and then land autonomously, all while gaining a better understanding of the technologies and infrastructure that will be required.
“We're using Project Bonsai because it allows us to quickly create and teach an AI just like if we were training a pilot on what to look for,” says Matt Holvey, senior manager of Intelligent Systems at Bell. “You can get the AI to understand what decisions to make about altitude and pitch based on the identified landing zone it sees.”
Unmanned aircraft cannot practice identifying landing zones and landing in the real world for three key reasons: safety concerns, potential damage to the aircraft and the amount of time it would take to train in enough real-world scenarios. That’s why AI practices in AirSim, a simulated environment, where it can execute thousands of landing identification scenarios in minutes. These exercises offer the experience needed for our aircraft to conduct safe operations and locate easy landing points to deliver goods and cargo to customers.
As the team think about the future of innovation and intelligent unmanned aircraft, Bell hopes to take on even more difficult challenges, such as AI-powered flight during inclement weather and low-visibility scenarios. For now, technology advancements at Bell continue to grow into new solutions that will unveil exciting new opportunities.
Comments