Perception-Aware Autonomous Exploration in Feature-Limited Environments

Abstract

Autonomous exploration in unknown environments typically relies on onboard state estimation for localisation and mapping. Existing exploration methods primarily maximise coverage efficiency, but often overlook that visual-inertial odometry (VIO) performance strongly depends on the availability of robust visual features. As a result, exploration policies can drive a robot into feature-sparse regions where tracking degrades, leading to odometry drift, corrupted maps, and mission failure. We propose a hierarchical perception-aware exploration framework for a stereo-equipped unmanned aerial vehicle (UAV) that explicitly couples exploration progress with feature observability. Our approach (i) associates each candidate frontier with an expected feature quality using a global feature map, and prioritises visually informative subgoals, and (ii) optimises a continuous yaw trajectory along the planned motion to maintain stable feature tracks. We evaluate our method in simulation across environments with varying texture levels and in real-world indoor experiments with largely textureless walls. Compared to baselines that ignore feature quality and/or do not optimise continuous yaw, our method maintains more reliable feature tracking, reduces odometry drift, and achieves on average 30\% higher coverage before the odometry error exceeds specified thresholds.

Figures (10)

• Figure 1: Our approach for perception-aware exploration applied on a UAV. (a) Odometry trajectory during an exploration task (blue arrows); arrow headings indicate the planned yaw angle. (b) Corresponding camera view. Our method maintains reliable feature tracking while successfully completing the exploration task.
• Figure 2: Our perception-aware exploration framework. Two key components are frontier selector and perception-aware trajectory optimisation.
• Figure 3: Our frontier selector. Frontier voxels are clustered by proximity, and candidate viewpoints are sampled from known free space to observe the frontier clusters. Visual features are overlaid, where blue intensity indicates feature quality: textured obstacles yield more and higher-quality features than textureless surfaces. Our selector prefers bottom-left viewpoints, which are farther from the current UAV position but provide stronger feature support than closer top-right viewpoints.
• Figure 4: Our yaw angle trajectory optimisation procedure. Dashed field of view sectors represent the waypoint yaws $\{\psi_j\}$ and arrows with different colours show covisible feature sets $\{\mathcal{F}_j\}$. The zoomed-in view shows the calculation of the relative angular velocity. For any point on trajectory, we decompose the velocity $\dot{\mathbf{p}_j}(t)$ in the direction of the feature vectors $\mathbf{f}_i - \mathbf{p}_j(t)$ and the perpendicular direction, expecting the yaw rate to compensate the relative angular velocity.
• Figure 5: Simulation environments with controlled texture richness. Left to right: Low-Texture (few textured pillars), Medium-Texture (partially textured construction site), and High-Texture (richly textured walls), each $10\,\mathrm{m}\times 10\,\mathrm{m}$.