PathGAN: Local Path Planning with Generative Adversarial Networks

07/08/2020
by   Dooseop Choi, et al.
0

Targeting autonomous driving without High-Definition maps, we present a model capable of generating multiple plausible paths from sensory inputs for autonomous vehicles. Our generative model comprises two neural networks, Feature Extraction Network (FEN) and Path Generation Network (PGN). FEN extracts meaningful features from input scene images while PGN generates multiple paths from the features given a driving intention and speed. To make paths generated by PGN both be plausible and match the intention, we introduce a discrimination network and train it with PGN under generative adversarial networks (GANs) framework. Besides, to further increase the accuracy and diversity of the generated paths, we encourage PGN to capture intentions hidden in the positions in the paths and let the discriminator evaluate how realistic the sequential intentions are. Finally, we introduce ETRIDriving, the dataset for autonomous driving where the recorded sensory data is labeled with discrete high-level driving actions, and demonstrate the-state-of-the-art performances of the proposed model on ETRIDriving in terms of the accuracy and diversity.

READ FULL TEXT

page 6

page 7

research
04/23/2018

Path Planning in Support of Smart Mobility Applications using Generative Adversarial Networks

This paper describes and evaluates the use of Generative Adversarial Net...
research
11/27/2016

SAD-GAN: Synthetic Autonomous Driving using Generative Adversarial Networks

Autonomous driving is one of the most recent topics of interest which is...
research
11/28/2022

SuperFusion: Multilevel LiDAR-Camera Fusion for Long-Range HD Map Generation and Prediction

High-definition (HD) semantic map generation of the environment is an es...
research
09/17/2019

Real-time Multi-target Path Prediction and Planning for Autonomous Driving aided by FCN

Real-time multi-target path planning is a key issue in the field of auto...
research
03/29/2018

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Understanding human motion behavior is critical for autonomous moving pl...
research
05/04/2019

SoilingNet: Soiling Detection on Automotive Surround-View Cameras

Cameras are an essential part of sensor suite in autonomous driving. Sur...
research
12/21/2018

Multi-Step Prediction of Occupancy Grid Maps with Recurrent Neural Networks

We investigate the multi-step prediction of the drivable space, represen...

Please sign up or login with your details

Forgot password? Click here to reset