Scaling Laws of Motion Forecasting and Planning -- A Technical Report

We study the empirical scaling laws of a family of encoder-decoder autoregressive transformer models on the task of joint motion forecasting and planning in the autonomous driving domain. Using a 500 thousand hours driving dataset, we demonstrate that, similar to language modeling, model performance improves as a power-law function of the total compute budget, and we observe a strong correlation between model training loss and model evaluation metrics. Most interestingly, closed-loop metrics also improve with scaling, which has important implications for the suitability of open-loop metrics for model development and hill climbing. We also study the optimal scaling of the number of transformer parameters and the training data size for a training compute-optimal model. We find that as the training compute budget grows, optimal scaling requires increasing the model size 1.5x as fast as the dataset size. We also study inference-time compute scaling, where we observe that sampling and clustering the output of smaller models makes them competitive with larger models, up to a crossover point beyond which a larger models becomes more inference-compute efficient. Overall, our experimental results demonstrate that optimizing the training and inference-time scaling properties of motion forecasting and planning models is a key lever for improving their performance to address a wide variety of driving scenarios. Finally, we briefly study the utility of training on general logged driving data of other agents to improve the performance of the ego-agent, an important research area to address the scarcity of robotics data for large capacity models training.

@article{baniodeh2025_2506.08228,
  title={ Scaling Laws of Motion Forecasting and Planning -- A Technical Report },
  author={ Mustafa Baniodeh and Kratarth Goel and Scott Ettinger and Carlos Fuertes and Ari Seff and Tim Shen and Cole Gulino and Chenjie Yang and Ghassen Jerfel and Dokook Choe and Rui Wang and Vinutha Kallem and Sergio Casas and Rami Al-Rfou and Benjamin Sapp and Dragomir Anguelov },
  journal={arXiv preprint arXiv:2506.08228},
  year={ 2025 }
}