Xbox Woodpecker Electrical Vehicle Interface

Oriekaose Chukwuyem Agholor and Andrew Pries presenting this project to Senator Gary Peters (on the far left) and National Cyber Director, Harry Coker, Jr. (on the right), during their visit to the University of Detroit Mercy

The System Block Diagram

Description

This project was part of our Hardware and Software Integration course and the VICEROY Scholar research program sponsored by the Department of Defense. It was oriented around using an Xbox controller to control the Woodpecker Electric Vehicle (EV) through a combination of RS232 USB serial communication, ROS2 publisher/subscriber nodes and CAN.

It pushed our minds and abilities to be able to create our own interfacing and electric vehicle control software that we can go on to improve upon and implement either in our own projects, the company we work for or our own company if we choose to found one. It is also a pathway into electrical vehicle cybersecurity as cars around 3 years back were being hacked through their CAN, compromising them.

The project involved installing ROS2 on two separate Raspberry Pi’s, creating a publisher C file on one Raspberry Pi, and a subscriber C file on the other. We then were able to connect the Xbox controller and receive data into the publisher using a format for event-driven architecture. We then manipulated the data in the way we wanted to send it to the subscriber.

Next, we were able to connect the car’s CAN bus to the CAN (Controller Area Network) shield on the subscriber Raspberry Pi. After that, we took the demo CAN send code provided and incorporated it into our ROS2 subscriber code that was receiving data from the topic and used it to send that data to the CAN of the vehicle.

Overall, the Xbox controller can forward steer, reverse steer, brake and accelerate/decelerate the vehicle by sending data to the modules responsible for each action. We used one joystick for steering and the other for braking. Also, we used each trigger for the forward and reversing respectively. We were able to get the steering and braking to work correctly, and although the code was written for acceleration, the state of the vehicle prevented us from testing it.

Set up

1. Install Ubuntu and Ros2 on both Raspberry Pis.
2. Install the CAN system on two Raspberry Pis using the instructions in RS485-CAN-HAT-user-manuakl-en.pdf.
3. Connect an RS485-CAN-HAT to each of the Raspberry Pis.
4. Integrate one of the Raspberry Pis with the Woodpecker Electric Vehicle (EV) platform by connecting the CAN High and CAN Low of the RS485-CAN-HAT to that of the EV platform.

Note:

• Ensure both Raspberry Pi's are on the same Ethernet or internal WI-FI network for the following steps.
• Make sure the ~/.bashrc file of both Raspberry Pi's have the same ROS_DOMAIN_ID and that ROS_LOCALHOST_ONLY is either set to 0 or deleted.

Programmed With

(ROS2)

Installation

There are 2 different folders containing the two major parts of the code for the project:

• Publisher_node

• Subscriber_node

As the names imply, the first folder contains the code for setting up and creating the publisher node to be installed on one Raspberry Pi, and the other folder contains the code for the subscriber node to be installed on the Raspberry Pi integrated with the EV platform.

Steps to install the program:

1. Create a ROS2 package on both Raspberry Pis by doing the following:

   a. Create a new folder, your workspace, on both devices and create a src folder in both.

   b. In the Bash terminals of both Raspberry Pis run the code below.

#Replace the address/yourWorkSpace/ with the src folder location
cd address/yourWorkSpace/src

#Creates a package
#Replace yourPackageName with any name of your choice
ros2 create pkg yourPackageName

3. Copy the subscribe.cpp file from the Subscriber_node folder and paste it in yourPackageName -> src of the Raspberry Pi (Subscriber node) integrated with the EV platform.
4. Copy the pubvelcpp.cpp file from the Publisher_node folder and paste it in yourPackageName -> src of the other Raspberry Pi (Publisher node).
5. Replace the CMakeLists.txt and package.xml files with the ones contained in the aforementioned folders for the specific nodes.
6. Run colcon build in the Bash terminal in your workspaces at the same level as the src folders you created initially.
7. Afterwards, execute source install/local_setup.bash in the install folders within the src folders.

Usage

To use an Xbox to control the WoodpeckervElectrical Vehicle Interface, do the following:

1. Run ls /dev/input/ in the bash terminal of the publisher node.
2. Connect the Xbox controller to the publisher node.
3. Run ls /dev/input/ again.
4. Compare both outputs and make note of the new input device file name that starts with event.
5. Run ros2 run yourPackageName pubvelcpp /dev/input/inputDeviceFileName in the Bash terminal in your workspace at the same level as the src folder.
6. Then, on your subscriber node run ros2 run yourPackageName subscribe in the Bash terminal in your workspace at the same level as the src folder.
7. Now, you can utilise the Xbox to control the EV platform using the button control layout provided below.

The Xbox Control Layout We Specified for Controlling the Electrical Vehicle Platform

Development Process

Check out the development process in my YouTube playlist

Presentation and Demo to Senator Gary Peters (on the far left) and National Cyber Director, Harry Coker, Jr.

Video Demo

Authors

This project was a result of the collaborative effort of:

1. Oriekaose Chukwuyem Agholor - Aspiring Robotics and Mechatronic Systems Engineer
2. Andrew Pries - Aspiring Computer Scientists
3. Brandom Smith - Aspiring Robotics and Mechatronic Systems Engineer
4. Edrees Ahmed- Aspiring Electrical Engineer

License

This project is licensed under the “Commons Clause” License Condition v1.0. See LICENSE for more information.

How to Contribute

Recommendations

• Utilise a more robust means of acceleration using a set of equations to convert the publisher data to correspond to hexadecimal values that are sent to the CAN bus for acceleration if possible.
• Allow other persons to view the project.
• Act on their feedback accordingly.