Abstract
We consider the task of visually estimating the relative pose of a drone racing gate in front of a nano-quadrotor, using a convolutional neural network pre-trained on simulated data to regress the gate's pose. Due to the sim-to-real gap, the pre-trained model underperforms in the real world and must be adapted to the target domain. We propose an unsupervised domain adaptation (UDA) approach using only real image sequences collected by the drone flying an arbitrary trajectory in front of a gate; sequences are annotated in a self-supervised fashion with the drone's odometry as measured by its onboard sensors. On this dataset, a state consistency loss enforces that two images acquired at different times yield pose predictions that are consistent with the drone's odometry. Results indicate that our approach outperforms other SoA UDA approaches, has a low mean absolute error in position (x=26, y=28, z=10 cm) and orientation ($\psi$=13${^{\circ}}$), an improvement of 40% in position and 37% in orientation over a baseline. The approach's effectiveness is appreciable with as few as 10 minutes of real-world flight data and yields models with an inference time of 30.4ms (33 fps) when deployed aboard the Crazyflie 2.1 Brushless nano-drone.
Abstract (translated)
我们考虑使用卷积神经网络（CNN）来估计纳米四旋翼无人机前方赛车门的相对姿态，该网络在模拟数据上进行了预训练以回归门的姿态。由于从仿真到现实之间的差距，预训练模型在实际世界中的表现不佳，必须适应目标领域。为此，我们提出了一种无监督域适应（UDA）方法，仅使用由无人机沿任意轨迹飞行并收集的实图序列；序列通过将无人机的里程计信息作为标注来自我监督地进行注释。在这个数据集上，状态一致性损失确保在不同时刻获取的两张图像所预测的姿态与无人机的里程计测量结果一致。实验结果显示，我们提出的方法优于其他最先进的UDA方法，在位置（x=26厘米, y=28厘米, z=10厘米）和姿态（$\psi$=13°）方面的平均绝对误差较低，相对于基线模型而言，在位置上提高了40%，在姿态上提高了37%。这种方法即使使用最少仅十分钟的现实飞行数据也能显示出显著效果，并且部署到Crazyflie 2.1无刷纳米无人机上的推理时间仅为30.4毫秒（每秒33帧）。
URL
https://arxiv.org/abs/2603.02936
