https://github.com/mrvplusone/steady

Simultaneous State Estimation and Dynamics Learning from Indirect Observations.
https://github.com/mrvplusone/steady

expectation-maximization-algorithm machine-learning particle-filters probabilistic-inference stochastic-dynamics

Last synced: 9 days ago
JSON representation

Simultaneous State Estimation and Dynamics Learning from Indirect Observations.

• Host: GitHub
• URL: https://github.com/mrvplusone/steady
• Owner: MrVPlusOne
• Created: 2021-07-29T04:57:05.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2023-04-04T04:43:42.000Z (over 1 year ago)
• Last Synced: 2024-05-22T00:04:34.331Z (6 months ago)
• Topics: expectation-maximization-algorithm, machine-learning, particle-filters, probabilistic-inference, stochastic-dynamics
• Language: Julia
• Homepage:
• Size: 19.8 MB
• Stars: 3
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# STEADY

_Learning stochastic dynamical systems from noisy and indirect obseravtions._



This repo contains the source code for the IROS 2022 paper [STEADY: Simultaneous State Estimation and Dynamics Learning from Indirect Observations](https://arxiv.org/abs/2203.01299). 

## Installing project dependencies

This project is written in [Julia](https://julialang.org/) and is tested on both MacOS and Ubuntu. We provide the script [install_linux.py](install_linux.py) to automatically install the binary dependencies on Linux.

When running the project for the first time, open the Julia REPL using the command `julia --project`.

Then, inside the REPL, type `>` followed by `instantiate` to let julia download all dependended packages and perform a precompilation.

Finally, create a file at `/configs/datadir.txt` and use its content to specify the location at which the trained models should be saved. e.g., 

```bash

mkdir configs; echo "data" > configs/datadir.txt

```

## Training kinodynamic models

To start training, inside the Julia REPL, first config the experiment

setting by defining a global variable named `script_args`:

```julia

script_args = (

    scenario=SEDL.RealCarScenario("alpha_truck"),

    train_method=:EM,

    gpu_id=0,

)

```

This specifies that we want to train on the ["alpha_truck" dataset](datasets/alpha_truck/) using expectation-maximization (i.e., the STEADY algorithm). `gpu_id=0` means that we want to use the first available GPU (set this to `nothing` if trained on CPU). You can find the full list of configurable parameters along with their acceptable values [here](https://github.com/MrVPlusOne/STEADY/blob/master/scripts/experiments/experiment_common.jl#L13).

We can then start the training with:

```julia

include("scripts/turn_off_displays.jl"); # this is only needed when a plot pane is not available. e.g., when running inside an ssh shell.

include("scripts/train_models.jl") # this starts the training

```

We also provide the following scripts to reproduce our paper results:

- [baseline_experiments.jl](scripts/experiments/baseline_experiments.jl): Corresponding to the "Main results" Section of the paper.

- [obs_noise_variation.jl](scripts/experiments/obs_noise_variation.jl): The "Impact of Observation Noise" Section.

- [particle_variation.jl](scripts/experiments/particle_variation.jl): The "Impact of Particle Quantity" Section.