# PITA: Physics-Informed Trajectory Autoencoder

arXiv: https://arxiv.org/abs/2403.11728 · HTML: https://mlsanity.com/paper/2403.11728/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 13 references that are in the map.
- First posted on arXiv 18 Mar 2024.

## Builds on (most cited first, 7 of 7)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [The highD Dataset: A Drone Dataset of Naturalistic Vehicle Trajectories on German Highways for Validation of Highly Automated Driving Systems](https://mlsanity.com/paper/1810.05642/) (2018)
- [INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps](https://mlsanity.com/paper/1910.03088/) (2019)
- [A Physics-Informed Deep Learning Paradigm for Car-Following Models](https://mlsanity.com/paper/2012.13376/) (2020)
- [Self-Supervised Action-Space Prediction for Automated Driving](https://mlsanity.com/paper/2109.10024/) (2021)
- [PIAug -- Physics Informed Augmentation for Learning Vehicle Dynamics for Off-Road Navigation](https://mlsanity.com/paper/2311.00815/) (2023)

## Built on by (most cited first, 2 of 2)

- [Physically-Grounded Goal Imagination: Physics-Informed Variational Autoencoder for Self-Supervised Reinforcement Learning](https://mlsanity.com/paper/2511.06745/) (2025)
- [Physics-Informed Tracking (PIT)](https://mlsanity.com/paper/2604.16895/) (2026)

## Abstract (arXiv)

Validating robotic systems in safety-critical appli-cations requires testing in many scenarios including rare edgecases that are unlikely to occur, requiring to complement real-world testing with testing in simulation. Generative models canbe used to augment real-world datasets with generated data toproduce edge case scenarios by sampling in a learned latentspace. Autoencoders can learn said latent representation for aspecific domain by learning to reconstruct the input data froma lower-dimensional intermediate representation. However, theresulting trajectories are not necessarily physically plausible, butinstead typically contain noise that is not present in the inputtrajectory. To resolve this issue, we propose the novel Physics-Informed Trajectory Autoencoder (PITA) architecture, whichincorporates a physical dynamics model into the loss functionof the autoencoder. This results in smooth trajectories that notonly reconstruct the input trajectory but also adhere to thephysical model. We evaluate PITA on a real-world dataset ofvehicle trajectories and compare its performance to a normalautoencoder and a state-of-the-art action-space autoencoder.