Wheel Ticks calculation code equation correction

[adnan-saood]

Description

I am trying to create my own odometry filter and I was using the simulation when I noticed that the ticks values were wrong.

In the code, file irobot_create_nodes\src\wheels_publisher.cpp, lines 79 and 82:

$$ \text{Ticks}_L = \frac{\theta_L}{2 \pi r} \times 508.8~~~~ \text{Ticks}_R = \frac{\theta_R}{2 \pi r} \times 508.8 $$

The correct formula is:

$$ \text{Ticks}_L = \frac{\theta_L}{2 \pi} \times 508.8~~~~ \text{Ticks}_R = \frac{\theta_R}{2 \pi} \times 508.8 $$

Without the $r$...

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)

How Has This Been Tested?

I ran the simulation and compared the wheels rotation reported by the dynamic_joint_states topic and the wheels rotation calculated by the formula:

$$ \theta_{L|R} = \frac{\text{Ticks}_{L|R}}{508.8} \times 2\pi $$

# Run this command (Gazebo)
ros2 launch irobot_create_gazebo_bringup create3_gazebo.launch.py 

# Or (Ignition)
ros2 launch irobot_create_ignition_bringup create3_ignition.launch.py 

Checklist

• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas (No need)
• I have made corresponding changes to the documentation (No need)

[alsora]

Hi @adnan-saood, thank you for your PR.
The first set of formulas that you wrote seem indeed wrong

$$ \text{Ticks}_L = \frac{\theta_L}{2 \pi r} \times 508.8~~~~ \text{Ticks}_R = \frac{\theta_R}{2 \pi r} \times 508.8 $$

If the wheel radius r is included in the formula, then the wheel rotation \theta should have been replaced with the linear distance travelled by the wheel d.

So the correct formulas are the following:

$$ \text{Ticks}_L = \frac{\theta_L}{2 \pi} \times 508.8 $$

or equivalently

$$ \text{Ticks}_L = \frac{d_L}{2 \pi r} \times 508.8 $$

After this premise, let's look at the simulation code.
The code is using get_dynamic_state_value("left_wheel_joint", "position"), so the question is whether the "position" of the joint represents d_L or \theta_L.
If it's d_L, then the formulas are correct, but if it's \theta_L, then they are wrong (and your PR fixes the issue).

I think that your PR is correct, but I don't know gazebo joints well enough to confirm this out of the top of my head.
Have you checked whether the returned value from that function is an angle or a linear distance?

[adnan-saood]

Hi @alsora, thank you for your prompt reply.

Yes I have confirmed that they are angular values since it is from the diff_drive_controller.
Also, I confirmed that the reported ticks for $N$ wheel rotation is round($N\times508.8$).