Loading [a11y]/accessibility-menu.js
Sensor fusion for functional safety of autonomous mobile robots in urban and industrial environments | IEEE Conference Publication | IEEE Xplore

Sensor fusion for functional safety of autonomous mobile robots in urban and industrial environments


Abstract:

To enhance productivity, assist workers and to have a more flexible material flow, autonomous mobile robots (AMRs) are used in industrial applications. To guarantee the s...Show More

Abstract:

To enhance productivity, assist workers and to have a more flexible material flow, autonomous mobile robots (AMRs) are used in industrial applications. To guarantee the safety of workers, most AMRs are equipped with safety laser scanners and stop if a person enters the protective field. To leverage the benefits of AMRs outdoors and in urban logistics, several new challenges arise. Severe weather conditions like rain, fog, snow, low lighting and direct sunlight restrict the capabilities of cameras in the visible spectrum, thermal cameras, laser scanners and time of flight cameras. Besides the safe detection of persons, the distance from the robot to the person has to be measured or estimated to provide a classification if the person intrudes the protective field of the AMR. The state of the art pedestrian detectors are based on neural networks and designed for advanced driver assistance systems. Therefore, most datasets for training the networks are acquired from the rooftop of cars and do not reflect the perspective and environment of AMRs. Our study reviews the state of the art sensors and pedestrian detection methods and shows the benefit of sensor fusion technologies based on artificial intelligence. Further, we discuss the limitations of these methods for industrial outdoor and urban AMR safety applications and propose methods to overcome these drawbacks.
Date of Conference: 06-09 September 2022
Date Added to IEEE Xplore: 25 October 2022
ISBN Information:
Conference Location: Stuttgart, Germany

References

References is not available for this document.