DOBB: Dynamic Oriented Bounding Boxes for Multi-Agent Collision Avoidance under State Uncertainty and Heterogeneous Dynamics
DOBB: Dynamic Oriented Bounding Boxes for Multi-Agent Collision Avoidance under State Uncertainty and Heterogeneous Dynamics
This research presents DOBB, an online geometric representation that converts current state estimates, uncertainty, and agent-specific response characteristics into a dynamic, direction-dependent safety margin. By deriving geometric features to characterize interactions between agents' DOBBs and using a reward function that preserves their physical interpretation, the method achieves a mean total failure rate of 2.6% across four evaluation environments. Performance remains robust under out-of-distribution shifts in actuation dynamics and state-estimation uncertainty, with collision-based failure rates staying at or below 2.3% in every tested condition.
The paper introduces DOBB, a geometric method for multi-agent collision avoidance that maps state estimates and uncertainty into dynamic safety margins, achieving a mean total failure rate of 2.6% across four environments while maintaining collision-based failures at or below 2.3% under out-of-distribution shifts in dynamics and state estimation.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported failure rate
2.6%
Reported trials
View original evidence
ures that jointly characterize the relative configuration and severity of interactions between agents' DOBBs, along with a reward function that preserves the physical interpretation of those interactions. Across four evaluation environments, DOBB achieves the lowest mean total failure rate of 2.6%. This performance persists under out-of-distribution shifts in actuation dynamics and state-estimation uncertainty, with
Open source S4
- paper: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ures that jointly characterize the relative configuration and severity of interactions between agents' DOBBs, along with a reward function that preserves the physical interpretation of those interactions. Across four evaluation environments, DOBB achieves the lowest mean total failure rate of 2.6%. This performance persists under out-of-distribution shifts in actuation dynamics and state-estimation uncertainty, with
Open source S4
collision-based failure remaining at or below 2.3% in every tested condition. We attribute this robustness to the geometric abstraction of the safety margin, which is less coupled to the specific uncertainty and actuation regimes encountered during training.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org