ostadabbas / SAVeS-Platform

An Evaluation Platform to Scope Performance of Synthetic Environments in Autonomous Ground Vehicles Simulation (ICASSP2023)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SAVeS: Scoping Autnonomous Vehicle Simulation Platform

This is a evaluation platform designed to help researchers to easily test and compare various datasets generated from simulation and their real-world counterparts on the selected algorithms. Currently, our platform supports the following geometric SLAM based algorithms and learning-based models via scripts and step-to-step guides:

Geometric-based Algorithms Learning-based Models
A-LOAM GeoNet
LeGO-LOAM AdaBins
ORBSLAM3 DPT

We provide scripts to process following datasets, with more supports coming along the way:

Real-world Datasets Synthetic Datasets
nuScenes GTA V - PreSIL
KITTI CARLA

We are also providing guides on evaluating the results. This paltform is under development.

About

An Evaluation Platform to Scope Performance of Synthetic Environments in Autonomous Ground Vehicles Simulation (ICASSP2023)


Languages

Language:Python 65.7%Language:MATLAB 34.3%