# TRACE: Ergodic Trajectory Optimization for Active Scene Reconstruction

arXiv: https://arxiv.org/abs/2608.02304 · HTML: https://mlsanity.com/paper/2608.02304/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 47 references that are in the map.
- First posted on arXiv 3 Aug 2026.

## Builds on (most cited first, 25 of 25)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [Habitat: A Platform for Embodied AI Research](https://mlsanity.com/paper/1904.01201/) (2019)
- [2D Gaussian Splatting for Geometrically Accurate Radiance Fields](https://mlsanity.com/paper/2403.17888/) (2024)
- [The Replica Dataset: A Digital Replica of Indoor Spaces](https://mlsanity.com/paper/1906.05797/) (2019)
- [ActiveNeRF: Learning where to See with Uncertainty Estimation](https://mlsanity.com/paper/2209.08546/) (2022)
- [NeU-NBV: Next Best View Planning Using Uncertainty Estimation in Image-Based Neural Rendering](https://mlsanity.com/paper/2303.01284/) (2023)
- [GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction](https://mlsanity.com/paper/2402.16174/) (2024)
- [ActiveGS: Active Scene Reconstruction Using Gaussian Splatting](https://mlsanity.com/paper/2412.17769/) (2024)
- [ActiveSplat: High-Fidelity Scene Reconstruction through Active Gaussian Splatting](https://mlsanity.com/paper/2410.21955/) (2024)
- [Active Implicit Object Reconstruction using Uncertainty-guided Next-Best-View Optimization](https://mlsanity.com/paper/2303.16739/) (2023)
- [NARUTO: Neural Active Reconstruction from Uncertain Target Observations](https://mlsanity.com/paper/2402.18771/) (2024)
- [Neural Visibility Field for Uncertainty-Driven Active Mapping](https://mlsanity.com/paper/2406.06948/) (2024)
- [ActiveGAMER: Active GAussian Mapping through Efficient Rendering](https://mlsanity.com/paper/2501.06897/) (2025)
- [POp-GS: Next Best View in 3D-Gaussian Splatting with P-Optimality](https://mlsanity.com/paper/2503.07819/) (2025)
- [GauSS-MI: Gaussian Splatting Shannon Mutual Information for Active 3D Reconstruction](https://mlsanity.com/paper/2504.21067/) (2025)
- [Affordance-Driven Next-Best-View Planning for Robotic Grasping](https://mlsanity.com/paper/2309.09556/) (2023)
- [Ergodic Exploration of Distributed Information](https://mlsanity.com/paper/1708.09352/) (2017)
- [Active Perception using Neural Radiance Fields](https://mlsanity.com/paper/2310.09892/) (2023)
- [Real-Time Area Coverage and Target Localization using Receding-Horizon Ergodic Exploration](https://mlsanity.com/paper/1708.08416/) (2017)
- [Ergodic Exploration using Tensor Train: Applications in Insertion Tasks](https://mlsanity.com/paper/2101.04428/) (2021)
- [Fast Ergodic Search with Kernel Functions](https://mlsanity.com/paper/2403.01536/) (2024)
- [Ergodic Coverage In Constrained Environments Using Stochastic Trajectory Optimization](https://mlsanity.com/paper/1707.04294/) (2017)

## Abstract (arXiv)

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.