CMTS: Conditional Multiple Trajectory Synthesizer for Generating Safety-critical Driving Scenarios
arXiv:1910.00099
Abstract
Naturalistic driving trajectories are crucial for the performance of autonomous driving algorithms. However, most of the data is collected in safe scenarios leading to the duplication of trajectories which are easy to be handled by currently developed algorithms. When considering safety, testing algorithms in near-miss scenarios that rarely show up in off-the-shelf datasets is a vital part of the evaluation. As a remedy, we propose a near-miss data synthesizing framework based on Variational Bayesian methods and term it as Conditional Multiple Trajectory Synthesizer (CMTS). We leverage a generative model conditioned on road maps to bridge safe and collision driving data by representing their distribution in the latent space. By sampling from the near-miss distribution, we can synthesize safety-critical data crucial for understanding traffic scenarios but not shown in neither the original dataset nor the collision dataset. Our experimental results demonstrate that the augmented dataset covers more kinds of driving scenarios, especially the near-miss ones, which help improve the trajectory prediction accuracy and the capability of dealing with risky driving scenarios.
Submitted to ICRA 2020, 8 pages, 7 figures
References in corpus (20)
- Neural Discrete Representation Learning
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- NICE: Non-linear Independent Components Estimation
- Interpolation Consistency Training for Semi-Supervised Learning
- Density estimation using Real NVP
- Domain Separation Networks
- MidiNet: A Convolutional Generative Adversarial Network for Symbolic-domain Music Generation
- BAGAN: Data Augmentation with Balancing GAN
- CAN: Creative Adversarial Networks, Generating "Art" by Learning About Styles and Deviating from Style Norms
- Argoverse: 3D Tracking and Forecasting with Rich Maps
- Joint Multimodal Learning with Deep Generative Models
- Interpreting the Latent Space of GANs for Semantic Face Editing
- Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions
- Interaction-aware Multi-agent Tracking and Probabilistic Behavior Prediction via Adversarial Learning
- A Closed-form Solution to Universal Style Transfer
- Predicting Vehicle Behaviors Over An Extended Horizon Using Behavior Interaction Network
- Online Vehicle Trajectory Prediction using Policy Anticipation Network and Optimization-based Context Reasoning
- Fully-Featured Attribute Transfer
- A Multi-Domain Feature Learning Method for Visual Place Recognition
- MTGAN: Speaker Verification through Multitasking Triplet Generative Adversarial Networks