Decentralized cooperative perception for autonomous vehicles: Learning to value the unknown

TL;DR

This paper introduces a decentralized, learning-based communication strategy for autonomous vehicles to efficiently share perception data, improving awareness while reducing communication overhead in cooperative perception scenarios.

Contribution

It proposes a novel DRL-based approach with a Locally Predictable VAE for selective information requests, enhancing perception accuracy with minimal data exchange.

Findings

Achieved 25% increase in complementary information with only 5% of perception requested.

LP-VAE outperforms state-of-the-art models in belief state prediction.

Effective in a CARLA driving simulator environment.

Abstract

Recently, we have been witnesses of accidents involving autonomous vehicles and their lack of sufficient information. One way to tackle this issue is to benefit from the perception of different view points, namely cooperative perception. We propose here a decentralized collaboration, i.e. peer-to-peer, in which the agents are active in their quest for full perception by asking for specific areas in their surroundings on which they would like to know more. Ultimately, we want to optimize a trade-off between the maximization of knowledge about moving objects and the minimization of the total volume of information received from others, to limit communication costs and message processing time. For this, we propose a way to learn a communication policy that reverses the usual communication paradigm by only requesting from other vehicles what is unknown to the ego-vehicle, instead of filtering on the sender side. We tested three different generative models to be taken as base for a Deep Reinforcement Learning (DRL) algorithm, and compared them to a broadcasting policy and a policy randomly selecting areas. In particular, we propose Locally Predictable VAE (LP-VAE), which appears to be producing better belief states for predictions than state-of-the-art models, both as a standalone model and in the context of DRL. Experiments were conducted in the driving simulator CARLA. Our best models reached on average a gain of 25% of the total complementary information, while only requesting about 5% of the ego-vehicle's perceptual field. This trade-off is adjustable through the interpretable hyperparameters of our reward function.