Published October 27, 2021 | Version v1
Conference paper Open

Robotic Infrared Vision: Enabling operation in fog and low-light environments

  • 1. German Aerospace Center (DLR) Institute for the Protection of Maritime Infrastructures, Bremerhaven, Germany

Description

The operation of robots in fog or smoke is a challenging task for both autonomous and remote-controlled operations. To increase the situational awareness and observational abilities, a legged robot was equipped with a thermal camera, which was integrated into the controlling unit to enable operation in environments which would not have been accessible without. The interface between the robot and thermal camera control will be addressed in this context, as well as the physical and mechanical setup. Additionally to the communication between Spot and the thermal camera, an option was developed to access the data stream from an external computer within the network. The software developed for this purpose will be introduced as well.
The performance of the system will be discussed on the basis of a field test in fog. Demonstrating that the implemented system enhanced the situational awareness of the operator and allowed for operation in a scenario, which would have been impossible for remote robot operation without the additional capability of thermal imaging. In particular, the system enabled the operator to detect a person in an environment with high fog density. The robot was further able to approach the person in an obstacle filled environment and therefore gather more detailed information about the scene.

Files

Robotic Infrared Vision - Enabling operation in fog and low-light environments.pdf

Additional details

References

  • Jong-Hwan Kim and Brian Y. Lattimer, Real-time probabilistic classification of fire and smoke using thermal imagery for intelligent firefighting robot Fire Safety Journal, 2015, volume 72, DOI:10.1016/j.firesaf.2015.02.007.
  • Alfred Weigert, Heinrich J Wendker, and Lutz Wisotzki, Astronomie und Astrophysik WileyVCH Verlag, 2005.
  • William C. Malm, Visibility Elsevier Science, 2016, page 37-41.
  • Houghton, H. G., The Size and Size Distribution of Fog Particles Physics, 1932, volume 2, pages 467-745 DOI:10.1063/1.1745072.
  • Boston Dynamics, Spot Specifications, [online accessed] 16.07.2021, URL: https://shop.bostondynamics.com/spot.
  • Intel, Intel Specifications NUC, 2021.
  • Infratec, InfraTec Specifications, [online accessed] 16.07.2021, URL: https://www.infratec.eu/thermography/infrared-camera/variocamhdx-head-600/.
  • Boston Dynamics, GXP Specifications, [online accessed] 16.07.2021, URL: https://shop.bostondynamics.com/spot-gxp
  • Ren, Shaoqing and He, Kaiming and Girshick, Ross and Sun, Jian, Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, volume 39, DOI:10.1109/tpami.2016.2577031.