Aggregated Sparse Attention for Steering Angle Prediction
This work addresses the problem of accurate steering prediction for autonomous vehicles, but it appears incremental as it adapts existing attention methods to a specific domain.
The paper tackles steering angle prediction in autonomous driving by applying a sparse attention mechanism to the visual domain and proposing an aggregated extension, showing improvements over models without attention and with different attention types.
In this paper, we apply the attention mechanism to autonomous driving for steering angle prediction. We propose the first model, applying the recently introduced sparse attention mechanism to visual domain, as well as the aggregated extension for this model. We show the improvement of the proposed method, comparing to no attention as well as to different types of attention.