PedCar

This package provides a implementation of an urban intersection scenario as a discrete state MDP. This model has been used in the following papers:

• M. Bouton, A. Nakhaei, K. Fujimura, and M. J. Kochenderfer, “Safe reinforcement learning with scene decomposition for navigating complex urban environments,” in IEEE Intelligent Vehicles Symposium (IV), 2019.
• M. Bouton, J. Karlsson, A. Nakhaei, K. Fujimura, M. J. Kochenderfer, and J. Tumova, “Reinforcement learning with probabilistic guarantees for autonomous driving,” in Workshop on Safety Risk and Uncertainty in Reinforcement Learning, Conference on Uncertainty in Artificial Intelligence (UAI), 2018.

Installation

For julia > 1.1, you can add the SISL registry and the JuliaPOMDP registry (recommended):

pkg> registry add https://github.com/sisl/Registry
pkg> registry add https://github.com/JuliaPOMDP/Registry
pkg> add https://github.com/MaximeBouton/PedCar.jl

otherwise you can install all the dependencies manually.

using Pkg
Pkg.add(PackageSpec(url="https://github.com/sisl/Vec.jl"))
Pkg.add(PackageSpec(url="https://github.com/sisl/Records.jl"))
Pkg.add(PackageSpec(url="https://github.com/sisl/AutomotiveDrivingModels.jl"))
Pkg.add(PackageSpec(url="https://github.com/sisl/AutoViz.jl"))
Pkg.add(PackageSpec(url="https://github.com/sisl/AutoUrban.jl"))
Pkg.add(PackageSpec(url="https://github.com/sisl/AutomotiveSensors.jl"))
Pkg.add("POMDPs")
using POMDPs
POMDPs.add_registry()
Pkg.add(PackageSpec(url="https://github.com/sisl/AutomotivePOMDPs.jl"))
Pkg.add(PackageSpec(url="https://github.com/MaximeBouton/PedCar.jl"))