For safe operation of the robot in the track area, it is necessary for the robot to be able to position itself with a high degree of accuracy and plan its path. In addition to sensors and GPS, computer vision is also used for this purpose.
Current driving tests show that the repetitive path accuracy is within a few centimeters.
The growth of vegetation on and around railroad tracks must be controlled regularly to ensure the safety, availability and longevity of the facilities. Today, herbicides are sometimes used for this purpose in order to get the fast and permanently growing grass and scrub under control.
Autonomous processes with robots are now used in agriculture, industry, logistics and many other areas. For the railroad sector, there are currently no marketable solutions that regularly cut back the vegetation in different environments and along sometimes very extensive routes.
To enable SBB to use robots to assist with maintenance next to the track, the project "Robot-based vegetation control (RoVeKo) in the track edge area" was launched. An interdisciplinary team from the fields of mechanical engineering, electrical engineering and IT is working together with SBB to achieve this goal.
Among other things, the project team is working with a self-developed robot platform according to the requirements of the railroads: the platform is to be
- be smaller and lighter than many devices used in agriculture, so that it can be transported easily
- to cope with the rough terrain,
- and to be able to work in the tight spaces along tracks.
- In addition, the robot must not violate the clearance gauge, a safety zone above the tracks. To do this, it must be able to locate itself very precisely and navigate with pinpoint accuracy.
Various implements were analyzed as part of the project. Bar and string mowers proved to be the most effective in the planned area of application. Both types of mower were adapted for the platform and tested for their advantages and disadvantages.
Currently, the team is working with various sensors and technologies to meet localization and navigation requirements. In addition, there are specifications for off-road capability, obstacle detection and communication.
Autonomous operation with the given safety requirements is another challenge. Current developments in battery capacity represent a major advantage for the robot, as the operating radius can be continuously increased. In the long term, this could be linked to solar-powered charging stations to achieve 24/7 availability in the vegetation months.
Link to the Competence Center (CC) responsible for the project: CC Mechanical Systems
