Neurosymbolic Planning Using Natural Language Communication for Cooperative Autonomous Vehicles
Neurosymbolic planning using natural language communication for cooperative autonomous vehicles
UNCAP is a new framework for cooperative autonomous driving that enables vehicles to communicate using natural language to share observations and reduce uncertainty. The approach uses vision-language models and uncertainty calibration to determine which messages to trust and how to integrate them into driving decisions. Evaluated in CARLA using the OPV2V benchmark, UNCAP shows significant improvements in communication efficiency, safety, and decision-making compared to traditional methods and prior language-based approaches.
The paper introduces UNCAP, a framework for cooperative autonomous driving that uses natural language communication to enhance situational awareness. It addresses the challenge of trust and relevance in multi-vehicle interactions by combining vision-language models with uncertainty-aware reasoning. The framework reduces communication bandwidth and improves safety metrics in simulated environments.
Source: Robohub · 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.
- Environment
- SimulationOpen source S13
Reported numbers
safety margin increase
4
View original evidence
hat blindly broadcast every observation to every other car, UNCAP consistently improved both efficiency and safety. Over various scenarios, our experiments yielded key qualitative gains of UNCAP over traditional cooperation schemes include: 63% less communication bandwidth 31% higher driving score 61% lower decision uncertainty 4× larger safety margins during near-miss events UNCAP avoids broadcasting every observati
Open source S15vision-language model performance
4
View original evidence
language descriptions first seems to act as a form of structured reasoning, distilling the essential information before it reaches the decision-making model. Equally encouraging, the improvements were consistent across several underlying vision-language models, including GPT-4o, GPT-4.1, GPT-4o-mini, and GPT-5. This suggests that the benefits stem from the communication strategy itself rather than from any particular
Open source S17model independence
4
View original evidence
language descriptions first seems to act as a form of structured reasoning, distilling the essential information before it reaches the decision-making model. Equally encouraging, the improvements were consistent across several underlying vision-language models, including GPT-4o, GPT-4.1, GPT-4o-mini, and GPT-5. This suggests that the benefits stem from the communication strategy itself rather than from any particular
Open source S17
- paper: not_reported
- paper: Reported pending Open source S21
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
driving decision Finally, the scene description generated by reasoning about all incoming messages is passed to an off-the-shelf vision-language model, which produces a driving decision such as merge, wait, or yield, along with a corresponding confidence score. What did this research teach us? We evaluated UNCAP in the popular CARLA driving simulator using the OPV2V benchmark, which is a widely used benchmark by auto
Open source S13
hat blindly broadcast every observation to every other car, UNCAP consistently improved both efficiency and safety. Over various scenarios, our experiments yielded key qualitative gains of UNCAP over traditional cooperation schemes include: 63% less communication bandwidth 31% higher driving score 61% lower decision uncertainty 4× larger safety margins during near-miss events UNCAP avoids broadcasting every observati
Open source S15
on to every nearby vehicle. It sharply reduces bandwidth while preserving the information needed for a driving decision. Perhaps the most surprising result was that sharing raw images actually performed worse than sharing language . Modern vision-language models appear to struggle when jointly reasoning over multiple camera feeds captured from different viewpoints. Translating those observations into concise natural-
Open source S16
language descriptions first seems to act as a form of structured reasoning, distilling the essential information before it reaches the decision-making model. Equally encouraging, the improvements were consistent across several underlying vision-language models, including GPT-4o, GPT-4.1, GPT-4o-mini, and GPT-5. This suggests that the benefits stem from the communication strategy itself rather than from any particular
Open source S17
shagra Gupta, Rohan Siva, Daniel Milan, Alexander T. Hogue, Sandeep P. Chinchali, David Fridovich-Keil, Zhangyang Wang, Ufuk Topcu . This work was nominated for the AAMAS 2026 best paper award.
Open source S21
Source：Robohub · robohub.org