Simultaneous Localization and Mapping (SLAM) algorithm for environment mapping using robot wheel odometry and LIDAR data
SLAM corrects for suboptimal robot dimension parameter settings
Occupancy Map - Odometry only, width = 500 mm | Path - Odometry only, width = 500 mm | Occupancy Map - SLAM, width = 500 mm | Path - SLAM, width = 500 mm |
---|---|---|---|
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
SLAM further refines robot trajectory and environment map when the robot dimension parameter is close to optimal
Occupancy Map - Odometry only, width = 730 mm | Path - Odometry only, width = 730 mm | Occupancy Map - SLAM, width = 730 mm | Path - SLAM, width = 730 mm |
---|---|---|---|
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Install micromamba or mamba as the package manager. To install micromamba, refer to the installation guide
To install the classifier:
- Clone the repo
git clone https://github.com/fyng/SLAM.git
cd SLAM
- Create virtual environment
micromamba env create -f environment.yml
micromamba activate slam
-
Create directory for data and plots
-
Run model In
main.py
, update the directory of the test folder. Chnage the car variables as needed
python main.py
ECE 5242 - Intelligent Autonomous Systems taught by Prof Daniel Lee & Travers Rhodes