Airbus completed its autonomous taxi, take-off and landing of an aircraft through fully automatic flight tests using onboard image recognition tech.
An autopilot is a system used to control the trajectory of a spacecraft, marine craft, or aircraft without the need for continuous manual control. However, autopilot has been around for the past few decades. The first aircraft autopilot was developed by Sperry Corporation in 1912. It was able to keep an aircraft straight and level and point the compass in a coherent direction, and to prove it, an aircraft made a spectacular public performance on the Seine outside Paris. Now, after so many years, Airbus has successfully completed its world-first Autonomous Taxi, Take-Off and Landing project.
The European manufacture has completed its autonomous taxi, take-off and landing of an aircraft through fully autonomous flight tests using onboard image recognition technology. The company conducted more than 500 test flights, of which around 450 flights were dedicated to collecting, supporting, and fine-tuning raw video data, while around six test flights, each with five take-offs and landing, were used to test the capabilities of autonomous flight. Now, Airbus can analyze the potential of these autonomous technologies to improve aircraft safety and enhance future aircraft operations.
The fully autonomous system depends heavily on machine learning and computer vision, uses a host of cameras, LiDAR, and radar, and has a seating capacity of more than 400 passengers. In January 2020, Airbus completed the first fully-automated take-off at Toulouse-Blagnac airport. In late June 2020, the company officially announced that the project was completed and ready for testing. The project was instigated to determine how self-ruling technologies such as automated tools and machine learning algorithms can be used for model creation, processing, and data labeling to help pilots focus more on mission management and strategic decision making, instead of focusing on aircraft operations.