Designing a Recurrent Neural Network to Learn a Motion Planner for High-Dimensional Inputs

TL;DR

This paper proposes a recurrent neural network model to improve motion planning in autonomous vehicles, aiming to bridge the gap between current rule-based methods and human-like driving behavior.

Contribution

It introduces a novel deep learning approach for motion planning in high-dimensional input spaces, serving as a baseline for future research in AV planning systems.

Findings

Demonstrates potential of RNNs for AV motion planning

Provides a baseline model for future research

Highlights advantages over rule-based methods

Abstract

The use of machine learning in the self-driving industry has boosted a number of recent advancements. In particular, the usage of large deep learning models in the perception and prediction stack have proved quite successful, but there still lacks significant literature on the use of machine learning in the planning stack. The current state of the art in the planning stack often relies on fast constrained optimization or rule-based approaches. Both of these techniques fail to address a significant number of fundamental problems that would allow the vehicle to operate more similarly to that of human drivers. In this paper, we attempt to design a basic deep learning system to approach this problem. Furthermore, the main underlying goal of this paper is to demonstrate the potential uses of machine learning in the planning stack for autonomous vehicles (AV) and provide a baseline work for ongoing and future research.