Conditional Vehicle Trajectories Prediction in CARLA Urban Environment

09/02/2019
by   Thibault Buhet, et al.
0

Imitation learning is becoming more and more successful for autonomous driving. End-to-end (raw signal to command) performs well on relatively simple tasks (lane keeping and navigation). Mid-to-mid (environment abstraction to mid-level trajectory representation) or direct perception (raw signal to performance) approaches strive to handle more complex, real life environment and tasks (e.g. complex intersection). In this work, we show that complex urban situations can be handled with raw signal input and mid-level representation. We build a hybrid end-to-mid approach predicting trajectories for neighbor vehicles and for the ego vehicle with a conditional navigation goal. We propose an original architecture inspired from social pooling LSTM taking low and mid level data as input and producing trajectories as polynomials of time. We introduce a label augmentation mechanism to get the level of generalization that is required to control a vehicle. The performance is evaluated on CARLA 0.8 benchmark, showing significant improvements over previously published state of the art.

READ FULL TEXT
research
06/18/2018

Conditional Affordance Learning for Driving in Urban Environments

Most existing approaches to autonomous driving fall into one of two cate...
research
03/09/2020

PLOP: Probabilistic poLynomial Objects trajectory Planning for autonomous driving

To navigate safely in an urban environment, an autonomous vehicle (ego v...
research
05/19/2018

End-to-end driving simulation via angle branched network

Imitation learning for end-to-end autonomous driving has drawn attention...
research
03/18/2023

Social Occlusion Inference with Vectorized Representation for Autonomous Driving

Autonomous vehicles must be capable of handling the occlusion of the env...
research
08/31/2021

Lane level context and hidden space characterization for autonomous driving

For an autonomous vehicle, situation understand-ing is a key capability ...
research
09/20/2019

Safer End-to-End Autonomous Driving via Conditional Imitation Learning and Command Augmentation

Imitation learning is a promising approach to end-to-end training of aut...

Please sign up or login with your details

Forgot password? Click here to reset