Challenges and implemented technologies used in autonomous drone racing
Autonomous drone racing (ADR) is a challenge for autonomous drones to navigate a cluttered indoor environment without relying on any external sensing in which all the sensing and computing must be done with onboard resources. Although no team could complete the whole racing track so far, most successful teams implemented waypoint tracking methods and robust visual recognition of the gates of distinct colors because the complete environmental information was given to participants before the events. In this paper, we introduce the purpose of ADR as a benchmark testing ground for autonomous drone technologies and analyze challenges and technologies used in the two previous ADRs held in IROS 2016 and IROS 2017. Five teams which participated in these events present their implemented technologies that cover modified ORB-SLAM, robust alignment method for waypoints deployment, sensor fusion for motion estimation, deep learning for gate detection and motion control, and stereo-vision for gate detection.
KeywordsAutonomous drone Drone racing Autonomous flight Autonomous navigation
J. Martinez-Carranza is thankful for the funding received by the Royal Society through the Newton Advanced Fellowship with reference NA140454. Team UZH thanks Elia Kaufmann, Antoni Rosinol Vidal, and Henri Rebecq for their great help in the software implementation and integration. Team of TU Delft would like to thank the organizers of the Autonomous Drone Race event. Team UNIST’s work was supported by NRF (2.180186.01 and 2.170511.01). All authors would like to thank the organizers of the Autonomous Drone Racing.
- 2.Brisset P, Drouin A, Gorraz M, Huard P-S, Tyler J (2006) The paparazzi solution. In: 2nd US-European competition and workshop on micro air vehicles (MAV)Google Scholar
- 4.Rojas-Perez LO, Martinez-Carranza J (2017) Metric monocular SLAM and colour segmentation for multiple obstacle avoidance in autonomous flight. In: IEEE 4th workshop on research, education and development of unmanned aerial systems (RED-UAS), OctoberGoogle Scholar
- 6.Foley JD, Van Dam A (1982) Fundamentals of interactive computer graphics. Addison-Wesley Longman Publishing Co., Inc., BostonGoogle Scholar
- 11.Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016) SSD: single shot multibox detector. In: European conference on computer vision (ECCV). Springer, pp 21–37Google Scholar
- 12.Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2015) Going deeper with convolutions. In: Computer vision and pattern recognition (CVPR)Google Scholar
- 13.Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: Advances in neural information processing systems (NIPS), pp 1097–1105Google Scholar
- 14.Szegedy C, Ioffe S, Vanhoucke V, Alemi AA (2017) Inception-v4, Inception-ResNet and the impact of residual connections on learning. In: AAAI conference on artificial intelligence (AAAI), pp 4278–4284Google Scholar