Organizations: Authors are with the Intelligent Robot Laboratory, University of Tsukuba, Japan. · Authors are with the Computer Vision Research Team, Artificial Intelligence Research Center (AIRC), National Institute of Advanced Industrial Science and Technology (AIST), Japan.
Exocentric cameras, mounted on walls or ceilings in indoor environments, offer wider vantage points and are less easily occluded than robot-mounted cameras, making them an appealing alternative for perception in robot navigation algorithms. Previous methods that use such cameras, however, require their pose and intrinsics to be known before deployment, demanding calibration that is time-consuming, requires specialized personnel, and must be repeated whenever a camera moves or a new room is navigated. We propose Robot-centric Obstacle Maps from Unknown and uncalibrated eXocentric cameras (ROMUX), an agent that navigates a robot toward point goals while avoiding obstacles using exocentric cameras that are both uncalibrated, meaning the system does not require their pose or intrinsic information as input, and unknown, meaning it is not fit to a specific camera during training and does not require pre-deployment fine-tuning. We train a cross-view transformer on large-scale simulation data to predict metric robot-centric obstacle maps from exocentric images, without conditioning image tokens on camera information. A small policy trained with Proximal Policy Optimization (PPO) then navigates on the predicted maps toward a goal coordinate relative to the robot. We benchmark our agent in simulation with cameras randomized per episode, and it matches the success rates of a PPO policy that navigates from a robot-mounted camera. Further, ROMUX works with one camera or multiple cameras with partial views and has been transferred zero-shot to a real robot using a camera at varying positions and orientations across various room layouts.
Figures & tables
Fig. 1: ROMUX trajectories on the real robot. Each column is one episode under a different camera position on the same obstacle layout.
Fig. 2: Simplified diagram of the ROMUX agent architecture.
Agent
Success
Collision
Time Limit
PPO No Image
25.80±2.55
74.19±2.61
0.01±0.06
PPO Fixed Exo
24.64±3.15
69.61±15.62
5.75±12.94
PPO Robot Cam
83.86±4.66
5.97±6.28
10.17±5.50
ROMUX 1-cam
85.04±2.42
9.87±0.72
5.09±1.72
ROMUX 2-cam
89.56±2.51
6.13±2.38
4.31±1.22
PPO GT Map
94.27±3.46
2.32±1.38
3.40±2.47
TABLE I: Navigation results in the cardboard rooms
Evaluation
Success
Collision
Time Limit
Known obstacles
86.40±2.94
8.92±4.82
4.68±2.17
Unknown obstacles
74.83±6.70
17.83±6.87
7.33±1.90
TABLE II: Navigation results with photorealistic furniture
Fig. 3: ROMUX trajectories in the 12×12 m rooms. Each row shows one episode, with the final image for each of the four cameras, and the robot’s trajectory overlaid on the ground-truth global obstacle map.
Fig. 4: ROMUX map estimation examples for unknown furniture in the photorealistic environment. Each column shows the two raw camera images and map prediction errors. False positives are orange and false negatives are green (for obstacles).
We study robot navigation in indoor environments from fixed exocentric cameras (infrastructure-mounted, external to the robot). We show that standard Proximal Policy Optimization (PPO), which succeeds with onboard cameras, fails to learn this navigation end-to-end in our setup. This failure results from the perspective mismatch between exocentric observations and egocentric action and goal formulations, which is too complex to resolve from the reward signal alone. To address this, we propose Viewpoint-Invariant Navigation from Exocentric Cameras (VINE). VINE uses an obstacle map estimation network, trained with supervised learning, that takes exocentric images, camera pose, robot pose, and room dimensions as input and produces a viewpoint-invariant obstacle map. This map can be used either by a PPO policy or by a planner such as A*. VINE's map estimation performs better than perspective mapping approaches, handles cameras placed across a wide range of positions and orientations without view-specific tuning, requires only approximate camera pose estimates (noise with a standard deviation of 30cm per axis in x, y, and z and about 20° of rotation error on average), is robust to intrinsic variation, and extends to two-camera setups with partial coverage. Despite operating from cameras randomized in position during evaluation, VINE surpasses the navigation success rate of PPO with an onboard camera in simulated environments.
Raul Steinmetz, Jair Augusto Bottega, Ayanori Yorozu +1
All authors are with the University of Tsukuba, Tsukuba, Ibaraki 305–8573, Japan.
Deploying navigation systems at scale requires a recipe that minimizes sensor assumptions, generalizes across robot embodiments, and trains efficiently. Yet, today's best systems depend on depth sensors, multi-camera rigs, or pre-built maps, limiting the hardware they support and increasing deployment cost. We introduce Robostral Navigate, an 8B vision-language model built around this scalability objective. The model consumes only a stream of monocular RGB images - the most ubiquitous sensor across robotic platforms and predicts waypoints by pointing to the next target location in the current camera view. Operating purely in image space, rather than robot-specific coordinates, makes the policy naturally robust to changes in camera intrinsics and scene scale, enabling deployment across wheeled, legged, and aerial robots without recalibration. We generate 2.4 million trajectories across 350k simulated scenes to reduce the reliance on real-world data collection and scale easily. We further introduce a prefix-caching training recipe that packs entire episodes into single training sequences, reducing training tokens by 22x and cutting training time from months to days. A tree-based attention mask prevents conditioning on previous ground-truth actions, encouraging visually grounded action prediction, and reinforcement learning is used to further improve exploration and recovery capabilities. On the Room-to-Room and Room-Across-Room in Continuous Environments (R2R-CE and RxR-CE) benchmarks, Robostral Navigate sets a new state of the art. On R2R-CE, it achieves a 77.4% success rate, surpassing the best monocular method by 10.5 points and the strongest depth- or multi-camera system by 5.3 points despite using only a single RGB camera. On RxR-CE, it reaches 75.1% success rate, outperforming all monocular baselines.
Visual navigation in unknown environments remains a core challenge in mobile robotics, especially for resource-constrained platforms. Most existing approaches rely on loosely coupled modular pipelines and strong assumptions on perception quality or environmental structure, often resorting to multi-modal sensor suites that increase system complexity and deployment cost. Vision-only navigation offers a lightweight alternative, but its performance degrades severely under motion blur, low texture, and illumination changes, largely because they neglect the tight coupling between commanded motion and perception. While perception-aware methods partially address this issue, they typically optimize individual modules and fail to propagate uncertainty consistently across the navigation stack. In this paper, we present UNSEEN, a unified uncertainty- and perception-aware navigation framework that explicitly couples localization, mapping, and planning using only a front-mounted camera. UNSEEN estimates sparse maps and robot poses with associated uncertainties at 6Hz, and leverages them to plan trajectories that jointly optimize task progress and estimation accuracy in receding-horizon. Simulations and extensive real-world experiments in unknown environments demonstrate the robustness of the proposed approach, with UNSEEN-SLAM reducing absolute translational error by 9.8% and UNSEEN-Plan improving estimation accuracy by up to 45% compared to state-of-the-art methods, while achieving a 100% task success rate.
Tommaso Faraci, Marco Camurri, Daniele Fontanelli +1