The drone outperformed traditional models’ tracking accuracy by 13.1 percent in outdoor experiments with dynamic flight paths and artificial impediments.
According to the team, the integration of bio-inspired mechanics with adaptive control and machine learning represents a notable advancement in drone agility and responsiveness under real-world conditions.
“The proposed data-driven approach could be further enhanced by incorporating efficient trajectory design, offering a promising avenue for future research,” said the team in the research paper.
The details of the POSTECH team’s research were published in the journal arXiv.