From Optical Flow to Safety: Visual Safety Certification via Scale-Free Relative Motion

Bazeela Banday; Jishnu Keshavan

cs.ROOct 15, 2026

From Optical Flow to Safety: Visual Safety Certification via Scale-Free Relative Motion

Authors: Bazeela Banday, Jishnu Keshavan

Organizations: Department of Mechanical Engineering, Indian Institute of Science.

Abstract

Safe autonomous navigation using monocular vision remains challenging because explicit obstacle states are not directly observable. This letter introduces a visual safety-certification paradigm that uses scale-free, depth-normalized relative motion encoded by monocular optical flow as the representation for collision-risk certification, without reconstructing obstacle position, velocity, geometry, or a metric world model. Unlike conventional optical-flow-based navigation approaches that rely on image motion to shape avoidance behavior, the proposed framework directly encodes collision-cone geometry and safety constraints in this relative-motion representation. An online Concurrent Learning observer estimates the normalized relative velocity and provides a computable estimation-error bound. This uncertainty is propagated into the collision cone through angular inflation and a monocular robot-footprint margin, yielding a Robust Vision-Based Collision Cone Control Barrier Function (RV-C $^3$ BF). An optimal-decay CBF quadratic program minimally modifies the nominal control input to enforce the resulting robust safety constraint under the stated assumptions. Extensive simulation and real-world experiments on a TurtleBot3 in cluttered environments with static and dynamic obstacles demonstrate collision avoidance using only monocular vision and onboard proprioception, with the estimation-error bound containing the measured error in $98.3\%$ of simulation samples and the robot achieving collision-free navigation in $93$ of $100$ real-world trials.

Figures & tables

Fig. 1: Monocular vision-based safety certification in real-world navigation. Collision geometry is constructed from scale-free relative motion obtained through monocular optical flow and onboard proprioception, without explicit obstacle-state reconstruction. (a) Navigation through a corridor with static obstacles. (b) Navigation among static obstacles and a moving obstacle. Yellow: robot; red: moving obstacle; square: start; star: goal.

Fig. 2: Monocular vision-based perception-to-control pipeline. De-rotated optical flow and onboard proprioception yield scale-free relative motion, estimated online using Concurrent Learning (CL). The nominal V-C 3 BF identifies the most critical collision cone, which is robustified using the CL estimation-error bound and monocular footprint margin to form the RV-C 3 BF. An optimal-decay CBF-QP enforces the resulting safety constraint. The resulting loop performs safety-critical navigation using monocular vision and onboard proprioception, with collision-cone safety certification performed directly in the native geometry of the sensing representation, without explicit reconstruction of obstacle position, velocity, or geometry.

Fig. 3: Mission-wide CL observer validation across three sequential static-obstacle encounters at

(2.5,0.2)

(5.5,0.2)

, and

(8.5,0.2)\,\mathrm{m}

. (a) Estimated

\hat{\xi}_{3}

(solid) versus ground truth (dashed). (b) In-plane estimation error and online bound

B_{c}(k)

(red, dashed). (c) Resulting trajectory through waypoints

(3.5,0.3)

(6.0,0.1)

, and

(9.5,0.4)\,\mathrm{m}

. Blue, orange, and green shading indicate obstacles 0, 1 and 2; gray dotted lines mark waypoint transitions.

Fig. 4: (a) Nominal V-C 3 BF vs. robust RV-C 3 BF escaping a 15-obstacle ring: the nominal controller collides in all three trials (red

\times

), the proposed method succeeds in all three. (b) 20 obstacles converging on the robot’s start position; five commanded goals, all reached collision-free. (c) Waypoint following through 60 moving obstacles via

(2.0,0.8)

and

(4.0,3.8)

to goal

(-2.0,-0.8)

. (d) Distances to all 60 obstacles (green) and to the nearest obstacle (black) in (c). The red solid curve shows the closest-approach obstacle, and the red dashed line marks the

0.30\,\mathrm{m}

collision radius.

Fig. 5: Snapshots of the waypoint-following mission of Fig. 4 (c). The purple trace shows the path traversed so far, colored by robot’s instantaneous speed ( $v$ ), illustrating the online adaptation of the RV-C 3 BF as it continuously deflects around moving obstacles while maintaining progress toward the active waypoint.

	Opt. Flow	Seg.	CL	Cone	CBF-QP	Ctrl.
Time (ms)	7.36	11.83	2.60	0.93	0.26	0.36

TABLE I: Average processing time of the vision and control stages of the implementation pipeline

Fig. 6: Experimental validation of the proposed RV-C 3 BF under different initial and final conditions, demonstrating collision avoidance and safe navigation in cluttered static and dynamic environments: a) open laboratory space, and b) narrow corridor.

Approach	Sensing	Safety	State/Map	Dynamic Val.
Optical flow [ 2 , 3 ]	Mono.	No	No	No
C 3 BF [ 5 ]	Full state	Yes	Required	Yes
RGB-D CBF [ 6 ]	RGB-D	Yes	Required	Yes ∗
Point-cloud CBF [ 7 ]	Depth/LiDAR	Yes	Required	Yes ∗∗
VLM-tuned CBF [ 8 ]	RGB ‡	Yes	Required	Yes ∗∗∗
Bioinspired SF [ 20 ]	Mono.	No	No	No

TABLE II: Comparison of representative approaches for vision-based collision avoidance and safety certification.

Robot Collision Avoidance

Robot collision avoidance develops sensing, planning, and control methods that keep robots safely separated from people, obstacles, and other robots.

See all 163 papers

Sep 29, 2026

Distilling Privileged Control Barrier Functions into RGB-Only Safety Filters for Dynamic Visual Navigation

RGB-only end-to-end visual navigation policies remain vulnerable to collisions in real-world dynamic environments, motivating a dedicated safety layer. Existing visual Control Barrier Function (CBF) approaches seek to provide safety from RGB observations, but often rely on real-time rendering or explicit scene reconstruction and are primarily designed for static scenes, limiting their practicality for onboard deployment. We propose a teacher-student visual distillation framework that transfers the safety behavior of a privileged CBF teacher to an RGB-only student filter for dynamic environments. The student maps a short RGB history, robot velocity, and a nominal control action directly to a safe action, while the teacher uses ground-truth robot and obstacle states in a real-to-sim dynamic Gaussian Splatting environment. To reduce the teacher-student information gap, the teacher constructs safety constraints only from obstacles observable within the student's RGB history. It also accounts for obstacle-velocity uncertainty to improve robustness to motion variations, while action augmentation exposes the student to diverse safe and unsafe nominal actions to better capture the safety boundary. At deployment, the student requires only RGB observations and robot velocity, without explicit 3D reconstruction or online rendering. Experiments show that the proposed method outperforms visual CBF baselines and improves the safety of RGB-based navigation policies under dynamic obstacle motion. Project page: https://syeon-yoo.github.io/distill-cbf-site/.

Seungyeon Yoo, Gawon Lee, Seungwoo Jung +2

Department of Aerospace Engineering, Seoul National University, Seoul, Republic of Korea

Control Barrier Functions Robotics Safe Robot Navigation Robot Safety Dynamic Obstacle Avoidance Vision-Based Robot Control+1

Ask PDF

Jun 10, 2026

SAFER-Nav: Enhancing Safety for Visual Robot Navigation via Segmentation-Aware Fine-Tuning

Vision-based navigation models, particularly foundation models, generate viable trajectories from RGB observations alone. However, even state-of-the-art transformer- and diffusion-based policies struggle to generalize in unfamiliar deployment environments containing unseen obstacles or shifted conditions. The resulting trajectories often remain goal-directed but unsafe. Existing efforts improve safety through external trajectory correction or internal geometric priors, yet the resulting policies are not trained to explicitly represent obstacle boundaries or traversable free-space structure. To address this, we propose a navigation model that incorporates these structures directly into the policy via fine-tuning and is designed to be compatible with diverse RGB-based backbones. Across multiple robot platforms, indoor environments, and static and dynamic obstacle scenarios, our method reduces collision frequency relative to ViNT, NoMaD, and their CARE-augmented variants while maintaining goal-reaching performance.

Geonyeong Ko, Giung Lee, Changjoo Nam

Dept. of Electronic Engineering, Sogang University, Seoul, Korea · Dept. of Computer Science, Rice University, Houston, TX, USA · Vertical Labs, Co., Ltd., Seoul, Korea

Robot Policy Learning Trajectory Generation Robot Navigation Safe Robot Navigation Robot Safety Visual Robot Navigation

Ask PDF

Aug 5, 2026

SCOPE: Field-of-View-Aware Path Planning in Unknown 3D Environments via Safety-Volume Certification

Safe navigation with a body-mounted limited-field-of-view sensor requires the complete robot-inflated volume of an intended motion to be observed and verified free before execution. We formulate this requirement as online safety-volume certification in an unknown voxel map and construct a certified graph whose vertices correspond exactly to positions with fully known-free safety volumes. Based on this representation, we propose SCOPE (Safety Certification through Observation Planning and Execution), a planning framework that decouples optimistic goal-directed guidance from certified execution. SCOPE converts the first uncertified point along an optimistic route into an explicit observation obligation, resolves it through target-centric viewpoint search, and recursively clears intermediate obligations when useful viewpoints are not yet certified-reachable. A certified preview mechanism and an observation-aware trajectory optimization backend enable smooth execution. We prove conditional complete planning: under ideal monotone sensing and exhaustive finite-domain graph search, SCOPE reaches the goal whenever a finite feasible sequence of certified sensing actions exists within its planning primitives. Across 60 randomized tasks in three unknown 3D environments, SCOPE reaches every goal while maintaining near-zero entry into non-certified inflated space. Preview reduces mean mission time by 27%, and real-robot demonstrations in two representative scenarios validate the complete system.

Junbin Yuan, Muqing Cao, Yunwoo Lee +2

Department of Mechanical Engineering at Carnegie Mellon University, Pittsburgh, PA 15213 USA · Robotics Institute, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213 USA · Department of Mechanical Engineering, Brigham Young University, Provo, UT 84602 USA

Constrained Motion Planning Safe Robot Navigation Robot Safety Graph-Based Planning Motion Planning

Ask PDF

Abstract

Robot Collision Avoidance

Explore similar work

Distilling Privileged Control Barrier Functions into RGB-Only Safety Filters for Dynamic Visual Navigation

SAFER-Nav: Enhancing Safety for Visual Robot Navigation via Segmentation-Aware Fine-Tuning

SCOPE: Field-of-View-Aware Path Planning in Unknown 3D Environments via Safety-Volume Certification