Robots: So that human-machine interaction does not become a brake on productivity

Many robot manufacturers now promise direct cooperation (collaboration) between humans and robots or at least barrier-free coexistence in the same workspace. In reality, however, this is associated with limitations: While the machines can work at high speed behind protective barriers, they have to significantly reduce their working speed in the presence of humans to protect them. This is at the expense of productivity, which is why many users ultimately continue to use traditional solutions. This is set to change soon.

As part of the European research project „Sharework“, a team from the Fraunhofer Institute for Machine Tools and Forming Technology IWU, together with six European research institutes and 13 companies, developed a speed controller and a comprehensive system for perceiving the robot environment. According to Fraunhofer IWU, this system also requires the robot's movement to be slowed down when approached by a human, albeit much less than before. The research team speaks of robot movements that are around 25 % faster than before. This would be a significant efficiency gain for users. The solution was trialled as part of the project at Spanish car manufacturer Seat, which belongs to Volkswagen AG.

Safe use of robots: Perception of the environment via camera and LiDAR

First of all, the reliable perception of the working environment had to be ensured. Cameras and LiDAR sensors (Light Detection and Ranging) are to be used for this purpose. The sensors recognise and categorise objects in a larger environment using pulsed laser light with a response time of 50 ms. The cameras, on the other hand, detect the near field with a response time of 10 ms. The higher working speed of the robot results from the combination of reaction times and monitoring ranges of both monitoring concepts.

The exact factor by which the process times can be reduced depends on the activity and the hardware used for robotics and LiDAR sensor technology, according to the researchers. According to the research team, customisation options for movement speed and acceleration increase acceptance among employees.

In addition to the hardware, software was also developed. Part of the software package is, for example, a module for assessing all safety aspects. This means that these can be taken into account at an early stage in the system planning, technically documented and digitally certified. At the same time, a digital image (digital twin) of the system with all relevant data is available for subsequent modifications.

Interaction with the heavy-duty robot

According to the researchers, the various software and hardware modules also enable heavy-duty robots in particular to interact with humans. Physical protective barriers such as fences can be dispensed with without compromising on safety. The Fraunhofer IWU's work focussed on overarching security aspects. The modular system is now able to understand a robot's environment and human actions. The system utilises sensor data and application knowledge, predictions of future conditions, intelligent data processing, augmented reality and gesture and speech recognition technology. In addition to applications in the automotive industry, the modules developed have already been trialled in the rail, metal and capital goods industries. It is also conceivable that they could be used in other industrial assembly and production processes.

Article from "vdi" from 17 March 2023