Reactive and Safe Road User Simulations using Neural Barrier Certificates
This work addresses safety and reactivity in traffic simulations for planning applications, but it appears incremental as it builds on existing imitation learning and control methods.
The paper tackled the problem of reactive and safe agent modeling for traffic simulators by proposing a model that learns high-level decisions from expert data and uses a low-level controller with learned barrier certificates. The result showed significant safety improvements over state-of-the-art methods, with smaller errors to expert data and better generalization to unseen traffic conditions.
Reactive and safe agent modelings are important for nowadays traffic simulator designs and safe planning applications. In this work, we proposed a reactive agent model which can ensure safety without comprising the original purposes, by learning only high-level decisions from expert data and a low-level decentralized controller guided by the jointly learned decentralized barrier certificates. Empirical results show that our learned road user simulation models can achieve a significant improvement in safety comparing to state-of-the-art imitation learning and pure control-based methods, while being similar to human agents by having smaller errors to the expert data. Moreover, our learned reactive agents are shown to generalize better to unseen traffic conditions, and react better to other road users and therefore can help understand challenging planning problems pragmatically.