Search papers, labs, and topics across Lattice.
This paper introduces Controllable Neural Variational Agents (CNeVA), a framework for simulating agents that not only mimic real-world traffic behaviors but also allow for controlled manipulation along interpretable axes. By employing a closed-form conjugate variational update and soft eligibility gates, CNeVA enhances the realism of simulated agents while maintaining a gradient signal for agents near reward thresholds. The results on the Waymo Open Motion Dataset show that CNeVA achieves competitive realism and controllability, outperforming traditional imitation models in terms of safety and steering responsiveness.
CNeVA reveals that smooth eligibility gates can significantly enhance agent controllability and realism in traffic simulations, outperforming traditional methods.
Realistic traffic simulation requires agents that imitate logged behavior and can also be steered along interpretable axes. Such controllability enables engineers to isolate variables, reproduce specific edge cases, and test autonomous systems without real-world risk. We introduce Controllable Neural Variational Agents (CNeVA), a controllable simulated-agent framework that learns to infer a per-agent Gaussian behavior latent from per-channel discounted returns via a closed-form conjugate variational update, conditioning a rectified-flow trajectory generator trained on a mixed channel-mask curriculum for classifier-free guidance. To tackle scarcity in reward signals, we propose soft eligibility gates that replace hard binary thresholds with smooth exponential decay, preserving the gradient signal for near-threshold agents. On the Waymo Open Motion Dataset, CNeVA attains competitive realism on the benchmark while exposing per-channel controllability that the higher-ranked imitation models lack. Speed- and acceleration-based steering produces monotone responses without stall-induced reward hacking. Safety controllability is monotone and substantial with the introduction of soft eligibility. We manage to achieve steerable map compliance under a context-residual return measure. Furthermore, our experiment demonstrates that steering metrics must be read alongside physical-plausibility guardrails to avoid reward-hacking confounds.
