Air Force Depot Pushes Multi-Robot Aircraft Inspection Workflow Forward
The Phase Three contract awarded by the Warner Robins Air Logistics Complex on-site demonstrates the relevance of a specific unmanned-systems problem: How to combine robots, autonomy software, and inspection data into a sustainable workflow? According to the terms of the award, Asylon is developing and demonstrating the MARIA multi-modal autonomous robotics for inspection of aircraft. In their own words, the company aims to provide an autonomous, repeatable, unattended general visual inspection capability for Department of the Air Force airframes. This development cannot be seen as merely as another robot being integrated into the maintenance facilities, as it involves connecting a drone and an inspection dog in a common autonomy platform and interface.
The Asylon solution features the company’s Guardian small uncrewed aircraft system, its DroneDog quadruped uncrewed ground vehicle, along with the company’s Range autonomy stack and DroneIQ command-and-control and data-visualization infrastructure. For depot maintenance organizations, this implies that the system is designed to conduct coordinated autonomous aircraft inspections and deliver the results via the same interface as opposed to making maintainers manage a set of disconnected robot feeds and data products.
As far as depot maintenance organizations go, the issue of integration is often more relevant than the robotic technology itself. Visual inspection of an aircraft is not about collecting images. It is about performing repeatable inspection, recording observations, storing telemetry and metadata, and providing the output in a form that allows maintainers to review the inspection quickly enough for the subsequent work on the line. The focus of MARIA is on the workflow level where automation should provide maintainers with decision support capabilities rather than merely collecting data.
For example, maintainers working with the Warner Robins systems will have to use DroneIQ software for the collation of imagery, LIDAR and point cloud data, telemetry, and metadata via both 2D and 3D views of inspection. The reason for this is the fact that visual inspection programs become more useful when the information is organized by the aircraft and the maintenance procedure rather than the sensor which collected the images. 3D view may allow maintaining aircraft geometry alignment, but 2D workflow is likely to match the existing review and documentation routine.
The Air Force approach to the program in the context of modernization and acceleration of maintenance is also noteworthy. In many cases, visual inspection of an aircraft is labor-intensive. Moreover, some manual visual inspection procedures present higher safety risks as they involve repeated proximity to large aircraft. The combined use of sUAS and UGV systems is presented by Asylon as a way to accelerate and make the process more consistent. The word consistent is important here as in the case of sustainment, repeatability is often more valuable than mere speed.
Finally, there is an issue of controls and operations of a multi-robot system which is worth mentioning. Synchronizing the activity of an aerial robot and a ground robot around aircraft structures in an actual depot is a complex task compared to operating a single drone. It includes mission orchestration, robust command-and-control, and the user experience that would not complicate the work of maintainers. The software solution that provides the perception of the multi-robot activity as the work of a single inspection system opens the door to the efficiency gains.
Warner Robins is a leader in aircraft maintenance and we are honored that the Air Force selected Asylon’s autonomous air and ground robotics platforms to enable faster, safer, and more consistent aircraft inspections, Anthony McCarty, Asylon’s senior vice president of government. Our solutions will give maintenance teams at the depot autonomy right at their fingertips to help improve readiness.
Asylon emphasized that the project is a flagship deployment at a major sustainment facility. According to the company, the effort could open up the pathways to the adoption of autonomous inspection systems in the broader defense and sustainment environments. The adoption potential is relevant only if the demonstration at the depot proves the efficiency of more than basic robot mobility. Namely, the key question here is whether a multi-modal system can produce inspection outputs in the format that can be used for review, recording, and decision-making without adding yet another technology stack.
This is why the program deserves attention from the unmanned-systems viewpoint. The crucial step is not just another robot in a hangar. The attempt is to combine the autonomy software, sensor data collection, and digital-twin visualization into the sustainable toolchain for maintainers.
By Stephen Wallace – Editor for AMI’s aerospace integration and unmanned mobility coverage.
