This repository contains C/C++ implementation of the grid-based 2D LiDAR SLAM, which were used in the following papers:
- A Unified Accelerator Design for LiDAR SLAM Algorithms for Low-end FPGAs (FPT, 2021)
- A Universal LiDAR SLAM Accelerator System on Low-Cost FPGA (IEEE Access, 2022)
The correlative scan matching (CSM) algorithm, which lies at the heart of the 2D LiDAR SLAM, is implemented on the TUL Pynq-Z2. The implementation of the custom CSM IP core is found in the following GitHub repository.
We used Xilinx Vivado HLS 2019.2 for synthesis and Vivado 2019.2 for the board-level implementation.
Our implementation is tested under the following environment:
- TUL Pynq-Z2
- Pynq Linux version 2.5 (based on Ubuntu 18.04)
- Python 3.6.5
- GCC 7.3.0
- Boost 1.65.1
- Eigen 3.3
Please refer the above papers if you find it interesting. Thank you!