Airbus Vision Landing Application Enables AI Autoland

Airbus has unveiled its Vision Landing Application, a computer vision and artificial intelligence system designed to enable fully automated landings at airports lacking traditional ground-based navigation infrastructure. Announced on June 10, 2026, ahead of the VivaTech forum in Paris, the technology represents the latest phase in the Smart Automation Roadmap of the manufacturer. The system utilizes onboard cameras and embedded AI to analyze runway features in real time, providing an independent positioning source for aircraft.

Advancing autonomous flight capabilities

The primary objective of the Vision Landing Application is to reduce reliance on external navigation aids like Instrument Landing Systems (ILS) or Ground Based Augmentation Systems (GBAS). In a press release detailing the technology, Airbus stated the goal is to create an independent positioning source to guide aircraft reliably.

“The goal of this research is to create an additional and independent positioning source to guide pilots and/or their aircraft reliably, opening up the perspective of bringing autoland (fully automated landing procedure) capabilities to airports that lack advanced ground infrastructure,” the company stated.

The current application builds on years of research conducted by Airbus and its innovation subsidiary, Airbus UpNext. The manufacturer launched the Autonomous Taxi, Take-Off & Landing (ATTOL) project on June 1, 2018, to test image recognition technology for airport navigation. This was followed by the DragonFly demonstrator project in November 2020, which focused on verifying operational relevance and scaling data processing for real-world complexities.

The Optimate demonstrator and embedded AI

The integration of these technologies is currently being tested through the Optimate demonstrator, launched by Airbus UpNext in 2023. Described as an “A350 cockpit on wheels,” the three-year research project will culminate in a complete automated gate-to-gate mission profile tested on an Airbus A350 flight test airframe.

A significant hurdle in deploying artificial intelligence in commercial aviation is regulatory certification. Airbus noted that AI in an aerospace context is constrained by strictly limited computing and power environments within the hardware of the aircraft.

“To design certifiable functions, Airbus engineers must fully master the hardware behaviour and maintain absolute visibility over all software lines of code,” the manufacturer noted.

Strategic AI partnerships

To accelerate its embedded AI capabilities, Airbus signed a partnership agreement with European artificial intelligence company Mistral AI on May 28, 2026. The collaboration focuses on deploying advanced AI across the commercial aircraft, helicopter, defence, and space divisions of the company.

A core component of the Mistral AI agreement is the development of “edge AI,” which involves deploying AI models directly on board aircraft for applications such as automatic object recognition. Catherine Jestin, Executive Vice President Digital at Airbus, stated the partnership paves the way for deploying high-impact use cases of trusted and responsible AI in aerospace.

Airbus will showcase the Vision Landing Application demonstration at the VivaTech forum in Paris from June 17 to June 20, 2026.

AirPro News analysis

We view the Vision Landing Application as a significant shift in how the aviation industry approaches all-weather operations. Historically, the burden of enabling automated landings fell on airport operators, requiring multimillion-dollar investments in ILS infrastructure and ongoing calibration. By shifting the technological capability to the airframe itself, Airbus is opening the door for airlines to operate reliably into smaller, less-equipped regional airports.

The certification of embedded AI for flight-critical phases like landing remains a formidable challenge. Traditional aviation software certification relies on deterministic outcomes, where a specific input always yields the exact same output. Machine learning models inherently challenge this paradigm. The explicit mention by Airbus regarding the need to maintain absolute visibility over all software lines of code indicates that the manufacturer is acutely aware of the regulatory hurdles ahead with agencies like the European Union Aviation Safety Agency (EASA) and the Federal Aviation Administration (FAA).

Sources: Airbus (Vision Landing Application)