I.AM: Impact aware manipulation in logistics

Alessandro Saccon, Associate Professor Robotics and Nonlinear Control (TU Eindhoven): “Now, we can make robots swiftly grab heavy objects with two arms, employing new methods that respect the natural impact dynamics of the system."

Impact
The research done by I.AM. resulted in robots that are impact-aware and are suited for fast physical interactions. That means the robots needed to learn to predict and react to what happens when they come into fast contact with heavy objects in the environment. That makes these robots different from traditional robots.

Typical robots are not designed to interact dynamically with their environment; making fast contact with surroundings is generally avoided. Much previous research has been focused on collision avoidance. The I.AM project targeted collision exploitation instead. The team looked into how a robot can, for example, hit heavy objects while ensuring that the execution of this action is reliable despite disturbances and perception inaccuracies.

The team discovered how to make robots swiftly grab a heavy object by developing new control and learning algorithms that respect the natural impact dynamics. They also understood how to use software simulations to obtain reliable predictions in intentional impact scenarios. This is essential in application domains requiring swift material handling, such as in warehouses and airports. The expected benefits include machine footprint reduction and limitation of worker injuries.

More info: https://i-am-project.eu/ 

Horizon 2020 - Industrial Leadership: Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)