Drive with the Flow

End-to-end autonomous driving systems have recently made rapid progress, thanks to simulators such as CARLA. They can drive without infraction of common driving rules on uncongested roads but are still struggling with dense traffic scenarios. We conjecture that this occurs because it lacks understanding of the dynamics of the surrounding vehicles, caused by the absence of explicit short-term memory within the perception path of end- to-end models. To address this challenge, we revise the perception module to explicitly model temporal information, by extending it with an auxiliary task that is well-known in computer vision research: optical flow. We generate a novel benchmark using the CARLA simulator to train our model, FlowFuser, and prove its superior ability to avoid collisions with other agents on the road.

Authors

• Enrico Mannocci
• Matteo Poggi
• Stefano Mattoccia

Keywords

• Training
• Computer vision
• Computational modeling
• Roads
• Buildings
• Proposals
• Vehicle dynamics
• Robotics and automation
• Optical flow
• Autonomous vehicles
• Collision
• Conjecture
• Temporal Information
• Auxiliary Task
• Computer Vision Research
• Perception Module
• Increase In The Number
• Neural Network
• Pedestrian
• Color Images
• Semantic Segmentation
• Consecutive Frames
• Data Frame
• Segmentation Map
• Gated Recurrent Unit
• Flow Map
• Optical Flow Estimation
• Planning Module