Knowing how to schedule tasks is a key knowledge crucial for developing low-power hardware setup with BigClown.

For this, it is important to be able to do event-driven programming easily. To achieve this, BigClown SDK provides tools for your application to be fully asynchronous. Key component is the Scheduler. Simply said, it holds pointers to functions with exact time when they should be run.


We've developed our own scheduler based on our need for simplicity, and low-power efficiency. It plans which task needs to be run and when. This scheduler is not the full-blown RTOS (Real Time Operating System) and it don't has the real co-operative multi-tasking. One task is run and when this task exits, then another task is run.

It is important to not block the task but do the necessary operation quickly and let the scheduler to run other tasks. In case you need to create some delay, then the one solution is to create for example a state machine and schedule to call the task later.

Get Time Since Start of the Program

Time since program started (timestamp, number of millisecond since start) can be retrieved by using bc_tick_t bc_tick_get() function.

Registering a Task

First thing you will probably run into when working with Scheduler is registering a simple task to be run in "x second(s) from now". Let's say, you have a function that will disable LCD display and do some other things when called.

static void disableLCD(void* param) {
    (void) param;
    // other things to do

Inside your code you will decide that you want to run this function in 5 second from now. This can be simply achieved by registering this function with bc_scheduler_register():

bc_scheduler_register(disableLCD, NULL, bc_tick_get() + 5000);

This function has a return type bc_scheduler_task_id_t. It returns ID of the task that had been registered by Scheduler. This ID can be used for unregistering from Scheduler.

Unregistering a Task

To unregister task from scheduler (for example when it is not needed to be run anymore) you have to use the void bc_scheduler_unregister(bc_scheduler_task_id_t task_id) function. This takes ID of the to-be-unregistered task as a parameter.

Planning to Run Registered Task

One time

To run registered task one more time in the future, you have to use one of these functions (taking task ID and number of milliseconds(ticks)) as parameters:

  • void bc_scheduler_plan_absolute
  • void bc_scheduler_plan_relative
  • void bc_scheduler_plan_from_now

To see full list of available functions, please see SDK documentation


To run task with ID 31415 in 5 seconds from now, you would simply call:

bc_scheduler_plan_from_now(31415, 5000);


Tasks can be run repeatedly, given absolute or relative time to be run. Absolute time says that task will be run in x seconds, no matter how long did the previous run of the function take. Relative time defines that task will be run exactly x seconds after this definition.


If task with ID 31415 is in fact the function disableLCD from our first example, we need to only add one line to this function. In this example, disableLCD will be always called exactly 2000 milliseconds after finishing previous run.

static void disableLCD(void* param) {
    (void) param;
    // other things to do
Chapters in this document:
Edit this page