From a03e17b75740cb45d370c6c1047de943e8c55829 Mon Sep 17 00:00:00 2001
From: Joshua Bird <40762456+jyjblrd@users.noreply.github.com>
Date: Fri, 24 May 2024 02:01:35 +0800
Subject: [PATCH] Update README.md

---
 README.md | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/README.md b/README.md
index 302a1d8..01b0026 100644
--- a/README.md
+++ b/README.md
@@ -1,4 +1,24 @@
 [Dissertation.pdf](https://jyjblrd.github.io/part_II_project_dissertation/Dissertation/dissertation.pdf)
 
+In this project, I:
+1. Design and implement a novel distributed monocular visual SLAM system, capable of
+localization, relative pose estimation, and collaborative mapping, all while being tolerant
+to degraded network conditions and not reliant on any single leader agent (section 3.2).
+2. Evaluate the performance of my system on standard datasets, **demonstrating its su-
+perior performance over comparable state-of-the-art systems (section 4.3).**
+3. Create a simulation environment for testing and evaluating my system locally (sec-
+tion 3.6).
+4. Develop a custom collision avoidance framework (section 3.3) and deploy it alongside
+my SLAM system on physical robots, **demonstrating the practical use cases of my
+system and benchmarking real-world performance (section 4.4).**
+5. **Contribute as a co-author to the paper The Cambridge RoboMaster: An Agile
+Multi-Robot Research Platform**. My distributed SLAM system is included in
+the paper and used to evaluate the robotics platform (section 3.7.1).
+6. Develop Multi-Agent EVO – the first open-source evaluation library for multi-agent SLAM
+systems (section 3.5).
+7. Develop the Raspberry Pi Video Publisher – a performant platform for SLAM data collec-
+tion and augmented reality visualizations – and set up a continuous integration and de-
+ployment pipeline to automatically deploy the latest builds to the devices (section 3.7.3).
+
 https://github.com/jyjblrd/distributed_visual_SLAM/assets/40762456/6db2d506-c0fd-4976-94fb-ae1da50cfd12