Percepto unveils Axion, a physical AI platform for integrating inspection data

Percepto unveiled Axion, a physical AI platform for integrating and analyzing inspection data, on October 27, 2026. The company says the platform includes more than 160 specialized AI models trained on more than 100,000 hours of autonomous inspection data. Through its new software offering, Axion can also handle data from drones and sensors made by other companies.

Axion is designed to ingest existing drone, sensor and inspection-record data, then analyze it in light of an asset’s history and condition. That approach addresses the challenge of operators having to manage inspection data separately for each piece of equipment. Percepto says customers can start with their existing inspection data, offering an analytics platform that does not require them to adopt the company’s own aircraft—a departure from traditional aircraft-centered operations.

Axion’s key distinction is that it separates inspection-data analysis from the company’s own aircraft, rather than focusing on improving drone performance. Percepto points to more than 160 models and more than 100,000 hours of inspection data as the platform’s foundation, while also proposing to ingest data from third-party drones, fixed and mobile sensors, and past inspection records. This suggests a shift toward a two-tier model: continuing its aircraft sales and operations business while offering software that handles inspection information regardless of the equipment used. The data flow is also significant. Inputs come from different devices and existing archives; the platform organizes them by asset and applies AI analysis based on asset history and condition; outputs include flagged issues and maintenance prioritization. The natural-language question feature is positioned as an interface to reduce the work of interpreting individual reports. However, this is the company’s description of the feature set; its announcement provides no performance figures showing how much inspection time or the rate of missed issues might improve. By positioning Axion as hardware-agnostic, Percepto broadens the competitive focus beyond drone specifications to include ingestion of diverse data, linking it to asset information, and designing workflows that connect analysis to maintenance decisions in the field. The ability to use existing equipment and records expands deployment options, while differences in data formats and asset identification will be an implementation focus. Percepto also named its Ranger and Harbor products as new data-collection options for Axion, suggesting a model in which the software serves as an entry point, while the company’s own hardware can be combined with it for additional data collection. The next points to examine are the range of supported third-party equipment and data formats, how AI models are updated and validated by experts, and the accuracy of analysis for different types of assets. The announcement also does not specify which records underpin natural-language responses or how field verification and corrections are incorporated into analysis. Without those details, it is difficult to assess whether the platform can provide the same level of decision support across different sites.

Axion is offered as software as a service (SaaS) and ingests data from supported third-party drones, fixed and mobile sensors, and existing inspection records. It organizes data by asset and interprets inspection results in light of asset conditions and historical trends. Users can also ask questions in natural language about assets and inspection results. Percepto also named its autonomous drone Ranger and autonomous operations station Harbor as new data-collection options for Axion. The company says it is working to consolidate engineering, manufacturing and assembly in the United States.