Posts by Collection

Due to the good maneuverability of the drone swarm, it shows high efficiency in the exploration of unknown environment, which can be applied in domains such as disaster relief and forest exploration. The method can allocate exploration regions to individual drones, while considering the uneven distribution of obstacles and calculating the local environmental complexity to adjust the exploration cost. All the code has been open-sourced at https://github.com/Jiang-Yufei/Multi-agent-exploration.git

A Self-Supervised Learning Approach with Differentiable Optimizationfor UAV Trajectory Planning

Research Project, Penn State University, 2024

We propose a self-supervised UAV trajectory planning pipeline that integrates a learning-based depth perception with differentiable trajectory optimization. A 3D cost map guides UAV behavior without expert demonstrations or human labels. Additionally, we incorporate a neural network-based time allocation strategy to improve the efficiency and optimality. The system thus combines robust learning-based perception with reliable physics-based optimization for improved generalizability and interpretability. Both simulation and real-world experiments validate our approach across various environments, demonstrating its effectiveness and robustness. Our method achieves a 31.33% improvement in position tracking error and 49.37% reduction in control effort compared to the state-of-the-art.

Yufei Jiang

Posts by Collection

portfolio

publications

talks

teaching