Stealth Blade Interferer

Project Overview

Stealth Blade Interferer(隐锋干扰者) is an autonomous driving car project that integrates multiple advanced systems, including LiDAR SLAM for navigation, YOLOv8 for object detection, and a Software-Defined Radio (SDR) jammer controlled by the car. The project employs two Raspberry Pi boards on the vehicle, along with two USRP devices deployed along the vehicle's path to simulate a wireless communication system for the car to interfere with. When the car detects the "xiaoba" object, it automatically triggers the SDR jammer.

System Components

1. LiDAR SLAM and Navigation:
   • LiDAR Model: Livox Mid-360, using the Livox ROS driver.
   • SLAM Algorithm: FAST-LIO2 (LiDAR-Inertial Odometry).
   • Platform: ROS Noetic.
   • Raspberry Pi 1: Handles SLAM, navigation, motor control, and manages the USRP device used as the jammer.
2. Object Detection (YOLOv8):
   • Raspberry Pi 2: Dedicated to running YOLOv8 for object detection due to its high computational demand.
   • Primary Goal: Continuously detects the "xiaoba" object in real-time.
3. Software-Defined Radio (SDR) Jammer:
   • Hardware: USRP device mounted on the car.
   • Control: Managed by Raspberry Pi 1.
   • Functionality: The vehicle interferes with a simple wireless communication system when it detects the "xiaoba" object.
4. Ground-Side Wireless Communication System:
   • USRP Devices: Two USRP devices are placed near the vehicle’s path.
   • Computer Interface: A computer connects both USRP devices to simulate a wireless communication system. The vehicle disrupts this communication once it detects the object.

System Architecture

• Raspberry Pi 1: Handles navigation, motor control, LiDAR SLAM, and the USRP SDR for the radio jammer.
• Raspberry Pi 2: Runs the YOLOv8 object detection model.
• Ground-Side USRP: The two USRP devices simulate the wireless communication system that the car disrupts once it detects the "xiaoba" object.

Setup Instructions

1. Raspberry Pi 1 Setup:
   • Install ROS Noetic, Livox ROS driver, and FAST-LIO2 for LiDAR SLAM.
   • Configure motor control and integrate the USRP SDR as a jammer.
2. Raspberry Pi 2 Setup:
   • Install and configure YOLOv8 for object detection.
3. Ground-Side Setup:
   • Connect the two USRP devices to a computer.
   • Set up the wireless communication system for the vehicle to interfere with.

Project Directory Structure

This project has two main components: the Car folder for navigation and object detection, and the USRP folder for communication and jamming.

Car/

• object-detection/:
  • Contains YOLOv8 model data (train/, test/, valid/) and trained weights under runs/detect/.
• ws_livox/src/:
  • chassis/: Controls motors and chassis operations.
  • FAST_LIO/ and FAST_LIO_LOCALIZATION/: SLAM and localization algorithms.
  • livox_ros_driver2/: LiDAR driver for Livox Mid-360.
  • donkey_nav/: Autonomous navigation scripts and launch files.

For your reference, the pointcloud file used in the demonstration is located at Car/ws_livox/src/FAST_LIO/PCD/jiao3map3/scans.pcd

USRP/

• communication/: GRC files (communicate1.grc and communicate2.grc) configure the SDR for communication.
• jammer/usrp_test1/: Jamming system code and build files for the SDR jammer.