BotLab (Fall 2023)

This project was developed for the ROB 550 course at the University of Michigan (Fall 2023). The focus of this lab is to build a full autonomy stack for a differential-drive robot using onboard sensors like LiDAR and odometry. The system includes modules for motion control, SLAM, path planning, and frontier-based exploration in a structured C++ framework.

---

Table of Contents

• Code Structure
• How to Build
• How to Run

---

Code Structure

🔧 Core Components

• src – All source code organized by modules:

  • mbot/: Motion controller implementation for differential-drive and omnidirectional robots.
  • slam/: SLAM components including particle filter, scan matching, and occupancy grid mapping.
  • planning/: Path planning and exploration algorithms including A* and frontier-based exploration.
  • utils/: Utility functions for geometry, timing, and LCM support.

• include – Corresponding headers for all modules, separated by namespace.

• assets – Contains visualizations and demo videos for checkpoint deliverables (e.g., FastRun.mp4, map.png).

• scripts – Installation and setup shell scripts.

  • install.sh: Installs dependencies and builds the workspace.

• services – Systemd service files for robot deployment:

  • mbot-slam.service, mbot-motion-controller.service

• build – Autogenerated folder by CMake after compilation (can be deleted and rebuilt).

• CMakeLists.txt: Project-level build file using CMake.

---

⚙️ How to Build

1. Install LCM (Lightweight Communications and Marshalling) if not already installed:

   sudo apt-get install liblcm-dev

2. Build the project using CMake:

   mkdir build && cd build
   cmake ..
   make -j4

---

🚀 How to Run

Launch SLAM

./build/mbot_slam

Launch Motion Controller

./build/mbot_motion_controller

Launch Exploration or Planning Server

./build/exploration
# or
./build/motion_planning_server

You can also set these up as systemd services using files in the services/ directory for auto-start on boot.

---

📦 Dependencies

• C++17
• CMake
• LCM
• Eigen3
• GStreamer

---

🧠 Concepts Practiced

• Odometry & motion models
• Particle Filter SLAM
• LiDAR-based mapping
• A* path planning
• Frontier-based exploration
• Modular C++ software architecture

---

👥 Contributors

• Trushant Adeshara
• Collaborators from ROB 550 (Fall 2023)

---

📜 License

This project is part of course-based work at the University of Michigan and follows the academic honor code for sharing.

---