Abstract

We propose a framework for active mapping and exploration that leverages Gaussian splatting for constructing dense maps. Further, we develop a GPU-accelerated motion planning algorithm that can exploit the Gaussian map for real-time navigation. The Gaussian map constructed onboard the robot is optimized for both photometric and geometric quality while enabling real-time situational awareness for autonomy. We show through viewpoint selection experiments that our method yields comparable Peak Signal-to-Noise Ratio (PSNR) and similar reconstruction error to state-of-the-art approaches, while being orders of magnitude faster to compute. In closed-loop physics-based simulation and real-world experiments, our algorithm achieves better map quality (at least 0.8 dB higher PSNR and more than 16% higher geometric reconstruction accuracy) than maps constructed by a state-of-the-art method, enabling semantic segmentation using off-the-shelf open-set models.

https://ieeexplore.ieee.org/abstract/document/11180885

Top: The Falcon 4 aerial platform is used for high-altitude experiments. The EvMAPPER sensor stack is mounted at the front. It is comprised of an IMU, a range sensor, an RGB camera, an event camera, and a synchronization board. Bottom left: artifacts of CMOS-based cameras in high-altitude photography: the sidewalk is washed out due to the limited dynamic range of the sensor. Bottom right: reconstructed frame with event cameras displaying a higher level of detail in challenging light conditions.