I have a question regarding the difference between GPS sensor and odometry. I know both of them give the robot’s coordinates, but one is simpler, and another is more complicated.
For programming skills, if we have GPS sensor, do we still need to use odometry? What are the pros and cons of both of them?
Hello! Most skills fields at competitions have GPS strips on the sides, but normal competitions usually don’t. It would really depend on what you are doing. If you already have an accurate autonomous in competitions, you should probably use only GPS for skills. But if you want to be precise with your turns in skills and your autonomous, then I would use both. Basically, GPS gives the robot a sense of where it is in the field, while odometry gives a sense of which direction the robot is facing.
Odometry can also be used to know where the robot is and can be more accurate in a match due to it not being reliant on a line of sight to the field perimeter.
While odometry can be used for rotation, that is not how most teams use it as the inertial sensor is more accurate. For position tracking, the gps sensor works fine but updates a lot slower and is less accurate then a well-made odometry system. I would recommend using odometry for position tracking over the gps sensor if you are able to build good tracking wheels and are confident with your programming skills.
Odometry is a method of measuring and calculating your robot’s absolute pose (position and heading). This is done by measuring the change in encoder rotation in the forward and sideways directions over time, and combining that with your robot’s heading to calculate a new position. This heading can be found using the IMU, or with multiple parallel encoders. Multiple parallel encoders is theoretically better, however is harder to implement (but it isn’t all that hard). I do not agree with @iseau395 ‘s notion that the IMU is more accurate.
The GPS sensor works by analyzing the gps tape around the field to get a good idea of what it’s position and heading are.
With well constructed tracker wheels, and well tuned code, an odometry system blows the GPS sensor out of the water. However, if you don’t think you will have time to learn about odometry, implement it, and tune it, the GPS sensor is not terrible. You will need to keep in mind that there is a dead zone once you get close to the field perimeter where the GPS sensor will have a harder time.