# Imitative Planning using Conditional Normalizing Flow

arXiv: https://arxiv.org/abs/2007.16162 · HTML: https://mlsanity.com/paper/2007.16162/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2020.

## Built on by (most cited first, 6 of 6)

- [Dual RL: Unification and New Methods for Reinforcement and Imitation Learning](https://mlsanity.com/paper/2302.08560/) (2023)
- [FloMo: Tractable Motion Prediction with Normalizing Flows](https://mlsanity.com/paper/2103.03614/) (2021)
- [Deep Generative Models for Offline Policy Learning: Tutorial, Survey, and Perspectives on Future Directions](https://mlsanity.com/paper/2402.13777/) (2024)
- [Learning Sampling Distributions for Model Predictive Control](https://mlsanity.com/paper/2212.02587/) (2022)
- [Individual Survival Curves with Conditional Normalizing Flows](https://mlsanity.com/paper/2107.12825/) (2021)
- [Sampling for Model Predictive Trajectory Planning in Autonomous Driving using Normalizing Flows](https://mlsanity.com/paper/2404.09657/) (2024)

## Abstract (arXiv)

A popular way to plan trajectories in dynamic urban scenarios for Autonomous Vehicles is to rely on explicitly specified and hand crafted cost functions, coupled with random sampling in the trajectory space to find the minimum cost trajectory. Such methods require a high number of samples to find a low-cost trajectory and might end up with a highly suboptimal trajectory given the planning time budget. We explore the application of normalizing flows for improving the performance of trajectory planning for autonomous vehicles (AVs). Our key insight is to learn a sampling policy in a low-dimensional latent space of expert-like trajectories, out of which the best sample is selected for execution. By modeling the trajectory planner's cost manifold as an energy function, we learn a scene conditioned mapping from the prior to a Boltzmann distribution over the AV control space. Finally, we demonstrate the effectiveness of our approach on real-world datasets over IL and hand-constructed trajectory sampling techniques.