Collision Detection and Reaction for Quadrotors Using Encoder-Integrated Tombo Propellers
This study presents a novel collision detection and reaction strategy for quadrotors equipped with deformable propellers (Tombo propellers). By integrating rotary encoders, we achieve real-time monitoring of individual propeller speeds, enabling rapid and reliable collision detection based on sudden velocity drops. The propeller speed variation during collisions is analyzed through simulations of the BLDC motor-Tombo propeller collision model. In addition, experimental validation is performed on a quadrotor to reinforce the correlation between propeller velocity changes and collisions. A collision detection algorithm based on this phenomenon successfully identifies eight out of twelve collisions during hover flights, while the remaining four minor collisions, which were undetected, had negligible effects on quadrotor stability. Subsequently, a collision response strategy that takes advantage of this algorithm for a quadrotor following a pre-defined trajectory is implemented. Experimental results demonstrate that this strategy is successfully applied to the Tombo quadrotor, whereas a quadrotor with rigid propellers fails due to propeller breakage on impact. This highlights the effectiveness of combining Tombo propellers for impact resilience with encoder-based velocity monitoring for reliable collision detection and response, ensuring safer quadrotor operations. (Video: https://www.youtube.com/watch?v=czliIvss5_I).
Authors
- Van Anh Ho (ORCID: https://orcid.org/0000-0002-4519-0448)
- Quan Khanh Luu (ORCID: https://orcid.org/0000-0002-8544-7260)
- Giuseppe Loianno (ORCID: https://orcid.org/0000-0002-3263-5401)
- Tien-Hung Pham (ORCID: https://orcid.org/0000-0003-0180-3676)
- Quang Ngoc Pham
Institutions
- Japan Advanced Institute of Science and Technology (JP)
- Purdue University West Lafayette (US)
- Robotics Research (United States) (US)
- New York University (US)
Publication Details
- Journal
- IEEE Access
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1109/access.2025.3572243
- Primary Topic
- Robotic Path Planning Algorithms
- Type
- article
- Field-Weighted Citation Impact
- 0.00