unizard
commented
6 years ago - This paper shows promising results for the ability of end to-end CNN frameworks to learn complex handling policies for self-driving vehicles.
- Using a ResNet18 model, they were able to achieve a strong prediction for optimal steering, throttle and braking values.
Model: ResNet18-variant
Dataset: Steering / Throttle / Braking
Institute: Stanford URL: https://arxiv.org/pdf/1806.11349.pdf Git: https://github.com/ymshao/End-to-End-Learning-for-Self-Driving-Cars, Window/Linux/Mac Interest: 3
Summary We introduce Ignition: an end-to-end neural network architecture for training unconstrained self-driving vehicles in simulated environments. The model is a ResNet-18 variant, which is fed in images from the front of a simulated F1 car, and outputs optimal labels for steering, throttle, braking. Importantly, we never explicitly train the model to detect road features like the outline of a track or distance to other cars; instead, we illustrate that these latent features can be automatically encapsulated by the network.