TRACE: Ergodic Trajectory Optimization for Active Scene Reconstruction
2026-08-03 • Robotics
Robotics
AI summaryⓘ
The authors point out that current 3D reconstruction methods pick the next best camera position step-by-step without considering the overall path, which can waste time moving between views. They propose treating the problem as an ergodic coverage task, where the camera's movement over time matches areas of high uncertainty and visibility in the scene. Their method updates the target areas dynamically and plans smoother, more efficient paths that balance exploring uncertain spots and covering the whole scene. Tested on a dataset, their approach improves image quality compared to older methods.
active reconstructionGaussian splattingnext-best-view (NBV)ergodic coveragesensor trajectoryuncertaintyvisibilitykernel-ergodic planningPSNRReplica dataset
Authors
Ziyue Zheng, Linli Shi, Bingkun He, Wen Jiang, Ziyun Wang
Abstract
Existing active reconstruction systems with Gaussian-splatting maps select observations greedily, optimizing a single next-best-view (NBV) at each step and connecting the chosen views by short-horizon path planning. This greedy decoupling disregards the global structure of scene information, producing inefficient trajectories that waste sensing capacity in transit between selected views. In this work, we study active reconstruction as an ergodic coverage problem: the time-averaged spatial statistics of the sensor trajectory should match a target information distribution induced by the current map. Our approach derives this target distribution online from uncertainty and visibility, and calculates ergodic trajectories via a kernel-ergodic horizon planner with gradient flow and footprint depletion, closing the loop between mapping and trajectory optimization. We thoroughly evaluate TRACE on the Replica dataset against the Next-Best-View (NBV) baselines, improving PSNR by 1.5 dB. Code: https://github.com/spikelab-jhu/trace-active-reconstruction.