Autonomous driving is the task of driving a vehicle without human conduction.
( Image credit: Exploring the Limitations of Behavior Cloning for Autonomous Driving )
A Simple and Versatile Framework for Object Detection and Instance Recognition
We present a traffic simulation named DeepTraffic where the planning systems for a subset of the vehicles are handled by a neural network as part of a model-free, off-policy reinforcement learning process.
While recent developments in autonomous vehicle (AV) technology highlight substantial progress, we lack tools for rigorous and scalable testing.
To our knowledge, this is the first successful case of driving policy trained by reinforcement learning that can adapt to real world driving data.
While most approaches to semantic reasoning have focused on improving performance, in this paper we argue that computational times are very important in order to enable real time applications such as autonomous driving.
In addition to requiring high accuracy to ensure safety, object detection for autonomous driving also requires real-time inference speed to guarantee prompt vehicle control, as well as small model size and energy efficiency to enable embedded system deployment.