Volume XLII-2/W6
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W6, 127-134, 2017
https://doi.org/10.5194/isprs-archives-XLII-2-W6-127-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W6, 127-134, 2017
https://doi.org/10.5194/isprs-archives-XLII-2-W6-127-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.

  23 Aug 2017

23 Aug 2017

OBSTACLE DETECTION SYSTEM INVOLVING FUSION OF MULTIPLE SENSOR TECHNOLOGIES

C. Giannì1, M. Balsi1, S. Esposito2, and P. Fallavollita1,2 C. Giannì et al.
  • 1DIET, Dept. of Information Engineering, Electronics and Communications, “La Sapienza” University of Rome, Italy
  • 2Oben s.r.l., Sassari, Italy

Keywords: Obstacle avoidance, UAV, Data fusion

Abstract. Obstacle detection is a fundamental task for Unmanned Aerial Vehicles (UAV) as a part of a Sense and Avoid system. In this study, we present a method of multi-sensor obstacle detection that demonstrated good results on different kind of obstacles. This method can be implemented on low-cost platforms involving a DSP or small FPGA. In this paper, we also present a study on the typical targets that can be tough to detect because of their characteristics of reflectivity, form factor, heterogeneity and show how data fusion can often overcome the limitations of each technology.