AI software helps machine-builders to develop adaptive picking robots
6 March 2025 
Siemens has developed machine vision AI software that allows machine-builders to develop their own AI-supported adaptive picking robots. The Simatic Robot Pick AI Pro software combines industrial AI technologies, software-defined controllers, and data-driven analytics to enable model-free, robot-based 3D picking of unknown objects.
The pre-trained deep-learning vision software works with individually adaptable vacuum multi-grippers. It delivers gripping poses (with six degrees of freedom) for a wide variety of items within milliseconds, regardless of their shape, size or packaging. Siemens says it will facilitate the development of cost-effective, autonomous, and scalable robot systems for single-piece order picking in sectors such as e-commerce. It also addresses labour shortages associated with monotonous picking tasks.
Robots equipped with the Siemens AI vision software can identify and handle a variety of unknown objects autonomously, ensuring increased flexibility and adaptability in dynamic environments. The software integrates with Siemens’ TIA platform, ensuring a continuous flow of data from order-picking cells to other operational processes.
Siemens will be demonstrating the new technology at this month’s Logimat intralogistics exhibition in Germany.
Simatic Robot Pick AI Pro is part of Siemens’ Industrial Operations X portfolio which combines software-defined automation and data-driven industrial systems. A key component of software-defined automation is Siemens’ Simatic AX development environment designed to make the creation and management of both physical and virtual controls more efficient. Virtual PLCs can provide more flexibility and scalability when deploying control systems as software containers based on industrial edge management systems.
Industrial Operations X integrates these technologies and supports collaboration between different systems, and the use of advanced technologies such as edge and cloud computing to improve operational processes. This allows machine-builders to develop flexible, scalable robot order-picking systems that can be adapted to meet specific requirements.
