This is a prototype of the program for the main computer. The real program is written in C and is currently still in construction. Although this is a prototype, it works exactly like the real version just slower. By natural of the language, python is slower than C but python is much faster to develop which is why it is common to prototype projects in python and write them later in C later. Python unfortunately has a feature called the GIL. This prevents python programs from running different threads parallel to each other on multiple CPU cores. This is also why the real program is being written in C. If the real version hasn't been finished yet and you need this program immediately then it will work.

The software is split into four packages: Communication, StateMachine, HealthMonitor, and HardwareControllers. The communication handles everything involving getting data from sensors to other services. StateMachine controls the state of the pod at all times. It is in charge of getting the pod from a braking position to full speed to braking again. The HealthMonitor checks to ensure that nothing is going wrong. If something goes wrong such as a software error or mechanical failure, the HealthMonitor will tell the StateMachine to stop the pod. Finally, the HardwareControllers are controlled by the StateMachine to control things like the motor and brakes.

Generally most configuration information and collection of glabal variables on the pod are stored in the config.ini file at the programs root. The two things not configured in the config.ini file are the logger and the network. The logger is configured using its own seperate log.ini file and the network configuration file will be created soon. To clarify, by the network configuration, I do not mean the configuration file for the tcp server, I am referring to the file that gets broadcasted across the network at the start up of the pod to let the microcontroller understand how the system looks.

Certain python 3 packages are needed to run some commands

~$ pip3 install -r requirements.txt

Command for final launch

~$ sudo python3 main.py

Soon this program will be turned into a service to begin at start up so this step won't be necessary in the final version.

While running, the pod software should present with a prompt. The prompt allows for a user remotely to control the pod's behavior manually. The manual control commands given to the outside user are:

• drift
  • Drifts the pod forward slowly
• launch
  • Launches pod at max speed
• estop
  • Emergency stops the pod
• shutdown
  • Shuts down pod and exits the program
• state
  • Displays current state of pod
• help
  • Displays usage information

Anyone using the software should consult the state diagram for the pod to understand how to control it properly. Many of the commands require the pod to be in a certain state before being used.

For questions about the system or the organization contact John Naylor at naylorjo@vcu.edu or jonaylor89@gmail.com