Autonomous driving maps extended beyond vehicle sight for safer navigation
Generation of Vectorized Maps Beyond Vehicle View
RoboticsArtificial IntelligenceMachine Learning
Summary
Autonomous cars use detailed maps to drive safely, but making these maps is expensive and they only cover what the car can currently see. The authors introduce a way to predict and draw what the map looks like beyond the car’s view, helping it plan better ahead. They created a new method called BeyondFormer and made a special dataset to test it. Their results show that the method can guess map details beyond the visible area, though there is still work needed to handle more complex driving situations.
autonomous drivinghigh definition mapsvectorized mapssensor field of viewmap generationmachine learningBeyondFormermap forecastingsafe navigationdataset
Authors
Clara Gomez, Alberto Jaenal, Antonio Artuñedo, Jorge Godoy, Jorge Villagra
Abstract
Autonomous driving relies on High Definition (HD) maps for safe navigation. Traditional HD maps construction is costly in hardware, data and human resources, which together with its update limitations hinders scalability. Recent works have proposed online alternatives for HD vectorized mapping from onboard sensors. However, sensor field of view is limited, and the range of the reconstructed maps ahead of the vehicle is insufficient for safe planning. This paper aims to address this limitation by proposing the novel beyond-view vectorized map generation problem: given vectorized maps of the area sensed by the vehicle (in-view), to generate plausible map continuations. To experimentally assess its feasibility, we propose BeyondFormer, which, to the best of out knowledge, is the first work designed towards beyond-view map generation. Given the novelty of the problem, we generate the first dataset specifically designed for it and evaluate the proposed approach. The results demonstrate consistent performance across diverse scenarios, establishing learning-based methods as a promising direction for map forecasting in autonomous driving. Beyond demonstrating the feasibility of the task, we provide an extensive discussion of the method's limitations and identify key future research directions for scaling it to more complex driving conditions. Code is available at https://git-autopia.car.upm-csic.es/beyondformer.