Vision Sensor

Note

For a full list of functions for interacting with the V5 Vision Sensor, see its C API and C++ API.

Setting Signatures

The first step to using the vision sensor is setting color signatures for the sensor to recognize as objects. This is done through the V5 Vision Utility program or programmatically.

C++
C
opcontrol.cpp
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
#define VISION_PORT 1
#define EXAMPLE_SIG 1

void opcontrol() {
  pros::Vision vision_sensor (VISION_PORT);

  pros::vision_signature_s_t RED_SIG =
    pros::Vision::signature_from_utility(EXAMPLE_SIG, 8973, 11143, 10058, -2119, -1053, -1586, 5.4, 0);


  vision_sensor.set_signature(EXAMPLE_SIG, &RED_SIG);
}
opcontrol.c
1
2
3
4
5
6
7
#define VISION_PORT 1

void opcontrol() {
  // values acquired from the vision utility
  vision_signature_s_t RED_SIG =
    vision_signature_from_utility(EXAMPLE_SIG, 8973, 11143, 10058, -2119, -1053, -1586, 5.4, 0);
}

Color Codes

Color codes can also be created from multiple signatures.

C++
C
opcontrol.cpp
1
2
3
4
5
6
7
8
#define VISION_PORT 1
#define EXAMPLE_SIG 1
#define OTHER_SIG 2

void opcontrol() {
  pros::Vision vision_sensor (VISION_PORT);
  pros::vision_color_code_t code1 = vision_sensor.create_color_code(EXAMPLE_SIG, OTHER_SIG);
}
opcontrol.c
1
2
3
4
5
6
7
#define VISION_PORT 1
#define EXAMPLE_SIG 1
#define OTHER_SIG 2

void opcontrol() {
  vision_color_code_t code1 = vision_create_color_code(VISION_PORT, EXAMPLE_SIG, OTHER_SIG);
}

Retrieving Objects

The primary function of the vision sensor is returning objects, or blobs of color detected by the sensor. The characteristics of an object are defined in vision_object_s_t.

By Size

The simplest way to interact with the vision sensor is to get an object by its size. 0 is the largest object detected by the sensor.

C++
C
opcontrol.cpp
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
#define VISION_PORT 1
#define EXAMPLE_SIG 1

void opcontrol() {
  pros::Vision vision_sensor (VISION_PORT);
  while (true) {
    vision_object_s_t rtn = vision_sensor.get_by_size(0);
    // Gets the largest object
    std::cout << "sig: " << rtn.signature;
    pros::delay(2);
  }
}
opcontrol.c
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
#define VISION_PORT 1

void opcontrol() {
  while (true) {
    vision_object_s_t rtn = vision_get_by_size(VISION_PORT, 0);
    // Gets the largest object
    printf("sig: %d", rtn.signature);
    delay(2);
  }
}

By Signature

If you have multiple signatures saved to the vision signature, you will most likely want to only look for objects of a particular signature. The get_by_sig() function implements this functionality.

C++
C
opcontrol.cpp
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
#define VISION_PORT 1
#define EXAMPLE_SIG 1

void opcontrol() {
  pros::Vision vision_sensor (VISION_PORT);
  while (true) {
    vision_object_s_t rtn = vision_sensor.get_by_sig(0, EXAMPLE_SIG);
    // Gets the largest object of the EXAMPLE_SIG signature
    std::cout << "sig: " << rtn.signature;
    // Prints "sig: 1"
    pros::delay(2);
  }
}
opcontrol.c
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
#define VISION_PORT 1
#define EXAMPLE_SIG 1

void opcontrol() {
  while (true) {
    vision_object_s_t rtn = vision_get_by_sig(VISION_PORT, 0, EXAMPLE_SIG);
    // Gets the largest object of the EXAMPLE_SIG signature
    printf("sig: %d", rtn.signature);
    // Prints "sig: 1"
    delay(2);
  }
}

Changing the Object Coordinates

Each returned object from the vision sensor comes with a set of coordinates telling where the object was found in the vision sensor’s field of view. The default behavior is to return the coordinates as a function of distance from the top left corner of the field of view - so positive y is downward and positive x is right. With the PROS API, you can change this behavior so that the center of the Field Of View is the (0,0) point for object coordinates. Positive y is still downward and positive x is still right, but negative y is upward of center and negative x is left of center in this configuration.

C++
C
initialize.cpp
1
2
3
4
5
#define VISION_PORT 1

void initialize() {
  pros::Vision vision_sensor (VISION_PORT, pros::c::E_VISION_ZERO_CENTER);
}
initialize.c
1
2
3
4
5
#define VISION_PORT 1

void initialize() {
  vision_set_zero_point(VISION_PORT, E_VISION_ZERO_CENTER);
}

Exposure Setting

In PROS Kernel 3.1.4 and earlier, the vision sensor exposure parameter was in the range [0,58]. In PROS Kernel 3.1.5 and newer, the parameter is scaled to be in the range [0,150] to match the Vision Sensor utility. As a result, there is a loss of information in this translation since multiple integers on the scale [0,150] map to the scale [0,58].