intelligent robotics:
the KI node in Munich

Modern times: In car production, robots work independently on car bodies; in power plants, they can reach inaccessible corners and carry out minor repairs. The use of machines in production and maintenance has long been common practice. Autonomous flying drones also fall into this category.

However, the majority of robots currently in use follow a strict set of rules. In a production line in the automotive industry, it is precisely defined which movements a machine must perform and which actions are set. If the situation changes abruptly, for example if a car body is placed on the conveyor in a different way, fixed programming can only deal with this situation to a limited extent.

Robots that can move on their own also follow a fixed route that must be entered beforehand. In case of a change in the route or if an obstacle blocks the passage, alternative routes must be defined. This is a problem with certain tasks, because it is not always possible to foresee all eventualities.

Autonomy through artificial intelligence

This is where artificial intelligence (AI) comes into its own. Because instead of individual steps, you can specify a goal for the robot. The machine recognizes the path independently – after it has gone through a training phase, of course. Based on the experience gained in the training, the robot can find and select the best method from the possible methods of achieving the goal. In addition, experience can also lead to the independent development of new methods. In the example of the maintenance robot in the power plant, which has to carry out a repair, this means independent route planning to the destination. AI in routing systems is not a new development. Existing navigation applications such as Google Maps have been using artificial intelligence for a long time.

Robotics from Munich

The centerpiece of the Digital Agenda Bavaria’s AI offensive is the node for artificial intelligence and robotics in Munich. At the Munich School of Robotics and Machine Intelligence (MIRMI), for example, researchers are working on intelligent interaction control for intuitive human-robot interaction. Here, MIRMI is working together with Rosenheim University of Applied Sciences, the company Franka Emika and the Center for Cognitive Sciences at the Technical University of Darmstadt on intuitive interaction between mostly elderly people potentially in need of care with care robots. MIRMI is running a similar project as part of the Geriatronic lighthouse initiative at the Garmisch-Partenkirchen site. A corresponding research center is being built there. In the field of intralogistics, the Munich node is involved in European projects such as the “Intra-Logistics with Integrated Automatic Deployment for safe and scalable fleets in shared spaces” (ILIAD). Here, the focus is on robotic systems in warehousing and internal logistics. The goal is to create automated fleets of heterogeneous robots that continuously optimize their operation and ensure efficient and safe operation in environments where humans also work.