Skip to content

Latest commit

 

History

History

task_planning

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 

Task Planning

Overview

High Level Overview

Task planning is the control center - the brains - of the robot. It operates at the highest level of the software subsystems and controls the overall actions of the robot: what tasks the robot attempts and in what order, where the robot moves, when the robot moves to the next task, etc. It uses data from sensor fusion and computer vision and gives instructions to controls. See the flow diagram in main README to see where it fits in the overall system.

Code Structure

The robot is controlled through running a series of tasks. Every task inherits from the Task object. The Task object's _on_task_start() function runs once when the task is run using Task.run(). Then _on_task_run() runs repeatedly until the task is finished using Task.finish(). If you have used Arduino or Processing before, this is analagous to the setup() and loop() methods.

Take a look at this example and its output below.

class ExampleTask(Task):
        
    def _on_task_start(self):
        print("This is an echo chamber:")
        self.count = 0
        
    def _on_task_run(self):
        print("echo")
        self.count += 1
        if self.count == 5:
            print("... you get the point")
            self.finish()

ExampleTask().run()

The result of running the above code would be:

This is an echo chamber:
echo
echo
echo
echo
echo
... you get the point

Prioritization of Tasks

Rationale (why did we design it this way)

Running Code

In Simulation

On the Robot

Guidelines for Writing Task Planning Code