OmniNav improves robot navigation in changing environments
OmniNav: Robust Long-Horizon Target Navigation in Dynamic Environments
Robotics
Summary
Robots must remember what they see, guess where targets are, and pick reachable spots to move to, even when things around them change. The authors present OmniNav, a system that keeps an updated 3D memory of the scene, updates beliefs about where targets might be using evidence from failed searches, and chooses navigation points that allow safe interaction. This lets robots handle situations where objects move or disappear and improves tasks like finding and picking up items. Experiments show OmniNav works better than earlier methods in various navigation and pick-and-place tests.
What this means in practice
- •For robotics engineers: Develop navigation systems that maintain accurate scene memory and update target location beliefs for robust robot operation in dynamic environments.
- •For warehouse automation teams: Enhance pick-and-place robots’ success rates by integrating feedback-aware navigation that adjusts to moved or missing objects.$Commercial implications: This paper enables improved real-world robotic picking systems for warehouses by increasing task completion reliability under changing item locations.
Authors
Yujie Tang, Meiling Wang, Jinhao Jiang, Sibo Zuo, Yinan Deng, Xinyu Zhang, Yufeng Yue
Abstract
Long-horizon target navigation requires a robot to sustain task execution across evolving observations, decisions, and physical interactions. This requires three coupled capabilities: maintaining valid scene memory, revising target beliefs under partial observability, and selecting interaction-feasible navigation endpoints. However, the state underlying each capability is only conditionally valid: scene representations become stale when objects move or disappear, unsuccessful searches alter beliefs over target locations, and geometrically convenient endpoints may still be infeasible for manipulation. To address these challenges, we present OmniNav, which formulates long-horizon navigation as continual inference over a factorized task state posterior coupling scene validity, target belief, and interaction feasibility. For representation, OmniNav incrementally constructs an updatable 3D object scene memory, preventing stale scene evidence from propagating to subsequent decisions. For exploration, it introduces an evidence-aware Bayesian belief-revision mechanism that derives dependency-aware region priors from semantic context, incorporates unsuccessful searches as negative evidence, and updates them for posterior-guided frontier selection. For interaction, OmniNav incorporates manipulation reachability and collision constraints into navigation-endpoint selection and propagates execution feedback through hierarchical closed-loop recovery. Extensive experiments demonstrate that OmniNav achieves the highest success rates among the compared methods on semantic ObjectNav and fine-grained instance navigation benchmarks, remains robust to target relocation, and improves real-world pick-and-place success from 53.3% to 71.7% over an adapted open-loop baseline. The project page of OmniNav is available at https://omni-nav.github.io/.