Design and implementation of a computer vision system for automated detection of aircraft on the apron. Development of algorithms for precise localization and position determination as a basis for autonomous control of a passenger boarding bridge.

Design and implementation of several features of an autonomous wood loading crane. Guidance of several developers during implementation. Support for commissioning on site in Finland.

Further development of a cloud-based solution for the efficient collection, management, and targeted selection of image data. Management of external annotation service providers. Creation of detailed annotation guidelines and training materials. Implementation of quality assurance concepts and consistency checks.

Design and development of a neural network for automated error detection in images. Training and optimization of the model using self-annotated data sets. Conducting systematic tests to validate detection accuracy. Development of post-processing algorithms for extracting statistical key figures (error type, defect size, frequency distribution)

Software development for autonomous timber loading cranes. From behavioral strategies and data structures to real-time execution in cognitive architectures. ?Hybrid? sensor processing using neural networks and classic image processing methods. Support for on-site commissioning in the USA.

Design and development of a system for grouping and classifying defects using clustering algorithms.