# A*SLAM: A Dual Fisheye Stereo Edge SLAM

**Authors:** Guoxuan Zhang  
**arXiv:** [1911.04063](https://arxiv.org/abs/1911.04063)  
**Date:** 2019-11-11  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1911.04063v1/

## Key points

- Guoxuan Zhang presents an innovative A*SLAM system utilizing dual fisheye stereo cameras
- System captures comprehensive view of environment by combining two sets of fisheye stereo cameras
- Key feature for SLAM is image edge, remaining robust in changing illumination conditions
- Panorama depth image extracted from each pair of fisheye stereo images to initialize SLAM feature
- Method involves using normal and inverted images interchangeably during edge-based SLAM process
- Improves accuracy, efficiency, and ensures robust performance in varying lighting conditions
- Offers a novel solution for edge-based SLAM using dual fisheye stereo cameras, advancing robotics and autonomous navigation systems

## Summary

Guoxuan Zhang presents an innovative A*SLAM system that utilizes dual fisheye stereo cameras to enhance simultaneous localization and mapping capabilities. This system captures a comprehensive view of the environment by combining two sets of fisheye stereo cameras. The key feature used for SLAM is the image edge, which remains robust even in changing illumination conditions. From each pair of fisheye stereo images, a panorama depth image can be extracted directly to initialize the SLAM feature. The proposed method involves utilizing both normal and inverted images interchangeably during the edge-based SLAM process. This not only improves accuracy and efficiency but also ensures robust performance in varying lighting conditions. Overall, this A*SLAM system offers a novel solution for edge-based SLAM using dual fisheye stereo cameras, paving the way for advancements in robotics and autonomous navigation systems.