# Quantity over Quality: Training an AV Motion Planner with Large Scale Commodity Vision Data

arXiv: https://arxiv.org/abs/2203.01681 · HTML: https://mlsanity.com/paper/2203.01681/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 35 references that are in the map.
- First posted on arXiv 3 Mar 2022.

## Builds on (most cited first, 30 of 30)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [CyCADA: Cycle-Consistent Adversarial Domain Adaptation](https://mlsanity.com/paper/1711.03213/) (2017)
- [End to End Learning for Self-Driving Cars](https://mlsanity.com/paper/1604.07316/) (2016)
- [VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation](https://mlsanity.com/paper/2005.04259/) (2020)
- [Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving](https://mlsanity.com/paper/1812.07179/) (2018)
- [ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst](https://mlsanity.com/paper/1812.03079/) (2018)
- [Categorical Depth Distribution Network for Monocular 3D Object Detection](https://mlsanity.com/paper/2103.01100/) (2021)
- [Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks](https://mlsanity.com/paper/1612.05424/) (2016)
- [A Survey of Motion Planning and Control Techniques for Self-driving Urban Vehicles](https://mlsanity.com/paper/1604.07446/) (2016)
- [End-to-end Interpretable Neural Motion Planner](https://mlsanity.com/paper/2101.06679/) (2021)
- [One Thousand and One Hours: Self-driving Motion Prediction Dataset](https://mlsanity.com/paper/2006.14480/) (2020)
- [Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Driving](https://mlsanity.com/paper/1906.06310/) (2019)
- [Stereo R-CNN based 3D Object Detection for Autonomous Driving](https://mlsanity.com/paper/1902.09738/) (2019)
- [A Survey of End-to-End Driving: Architectures and Training Methods](https://mlsanity.com/paper/2003.06404/) (2020)
- [Urban Driver: Learning to Drive from Real-world Demonstrations Using Policy Gradients](https://mlsanity.com/paper/2109.13333/) (2021)
- [DSGN: Deep Stereo Geometry Network for 3D Object Detection](https://mlsanity.com/paper/2001.03398/) (2020)
- [Driving Policy Transfer via Modularity and Abstraction](https://mlsanity.com/paper/1804.09364/) (2018)
- [LIGA-Stereo: Learning LiDAR Geometry Aware Representations for Stereo-based 3D Detector](https://mlsanity.com/paper/2108.08258/) (2021)
- [AFDet: Anchor Free One Stage 3D Object Detection](https://mlsanity.com/paper/2006.12671/) (2020)
- [Learning to Evaluate Perception Models Using Planner-Centric Metrics](https://mlsanity.com/paper/2004.08745/) (2020)
- [Learning Linear Transformations for Fast Arbitrary Style Transfer](https://mlsanity.com/paper/1808.04537/) (2018)
- [Autonomy 2.0: Why is self-driving always 5 years away?](https://mlsanity.com/paper/2107.08142/) (2021)
- [Image Question Answering using Convolutional Neural Network with Dynamic Parameter Prediction](https://mlsanity.com/paper/1511.05756/) (2015)
- [What data do we need for training an AV motion planner?](https://mlsanity.com/paper/2105.12337/) (2021)
- [Testing the Safety of Self-driving Vehicles by Simulating Perception and Prediction](https://mlsanity.com/paper/2008.06020/) (2020)
- [Collaborative Augmented Reality on Smartphones via Life-long City-scale Maps](https://mlsanity.com/paper/2011.05370/) (2020)
- [Scalable Primitives for Generalized Sensor Fusion in Autonomous Vehicles](https://mlsanity.com/paper/2112.00219/) (2021)

## Abstract (arXiv)

With the Autonomous Vehicle (AV) industry shifting towards machine-learned approaches for motion planning, the performance of self-driving systems is starting to rely heavily on large quantities of expert driving demonstrations. However, collecting this demonstration data typically involves expensive HD sensor suites (LiDAR + RADAR + cameras), which quickly becomes financially infeasible at the scales required. This motivates the use of commodity sensors like cameras for data collection, which are an order of magnitude cheaper than HD sensor suites, but offer lower fidelity. Leveraging these sensors for training an AV motion planner opens a financially viable path to observe the `long tail' of driving events. As our main contribution we show it is possible to train a high-performance motion planner using commodity vision data which outperforms planners trained on HD-sensor data for a fraction of the cost. To the best of our knowledge, we are the first to demonstrate this using real-world data. We compare the performance of the autonomy system on these two different sensor configurations, and show that we can compensate for the lower sensor fidelity by means of increased quantity: a planner trained on 100h of commodity vision data outperforms the one with 25h of expensive HD data. We also share the engineering challenges we had to tackle to make this work.