Interceptor 4.0 - Team Delamain

Team Members:

• Zhongheng (Heng) Li aka heng2j
• Nicolas Morant aka DevGlitch
• Huayu (Jack) Tsu aka codejacktsu

Description

We wanted to build a level 4 autonomous law enforcement vehicle, Interceptor 4.0, that patrols neighborhoods and when needed chases vehicles.

The challenges Interceptor 4.0 would face are as follows: lane detection and tracking, object detection and avoidance, GPS navigation, self-parking, specific vehicle tracking, and car chasing.

Project Report

https://github.com/heng2j/delamain/blob/master/Project_Report/Project_Report_Team_Delamain.pdf

Presentation

https://youtu.be/7PMrhMN3heU

Demos

https://github.com/heng2j/delamain/tree/master/demos

Getting Started

Follow the below instructions in order to run Interceptor 4.0 on your machine.

Prerequisites

• CARLA v0.9.10
  https://github.com/carla-simulator/carla
  This project was developed using this specific version of CARLA.
  We cannot guarantee that it would work with any higher version.

Installation

1. Clone the repo

   git clone https://github.com/heng2j/delamain.git

2. Install any missing packages.
   • We recommend using a conda environment with Python 3.7.

Running

1. Start Server Map: Open CARLAUE4.exe

2. Load the map that you'd like. By default CARLA loads map Town03

3. Run any of the command below.
   Please make sure you are in the correct directory.

• Base Model 3

Third iteration of our base model with full capabilities.
Press "i" to activate GPS from current location to destination.
Press "p" to toggle autopilot
Press "h" to toggle hotkey map

   python base_model3.py

• GPS Navigation Base Model

This file gives you a short demo of the navigation system. You can change the destination location. Please ensure the destination is for the load CARLA map.

   python base_model_nav.py

• Self-Parking Base Model

This file gives you a short demo of the self-parking feature. In this file you have the option to change from perpendicular to parallel parking.

   python base_model_park.py

• Road Network Map

This script enables you to visualize the road network of any CARLA map.

   python road_network_map.py

• Spectator Location

This script gave you the ability to get the exact location of the spectator view. It gives you the location in CARLA Location (x, y, z).

   python spectator_location.py

• Topology Edge & Node

This script enables you to store the topology data (edges and nodes) in two parquet files. These files are in a format that enables you to use with Network X.

   python topology_edge_and_node.py.py

• Topology Waypoints Visualizer

This script is a visualizer in CARLA of the topology waypoints. Make sure to run the previous script in order for this one to work.

   python topology_waypoints_visualizer.py

• Base Model 3 Car Chasing

Drive as leading car

   python base_model3_car_chasing.py

• Car Chase Demo v2

Run as chasing car

   python car_chase_demo_v2.py

• Car Chase Demo v3

Run car chasing with dynamic Frenet short-term trajectory planning

   python car_chase_demo_v3.py

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Zhongheng (Heng) Li - Github

Nicolas Morant - Personal Website & Github

Huayu (Jack) Tsu - Github

Project Link: https://github.com/heng2j/delamain