Answers

8.1 See as numbers

Learning · Robot club · about 15 min

What this lesson is about

What a model gets: the sensor views as lists of numbers.

• The depth grid is 64 numbers
• The level rows
• Same wall, different numbers
• A picture is numbers too

Questions 7 marks in all

1. [1 mark]How many numbers are in the depth grid from tof_grid()?

   Answer: 64. Eight rows of eight: 64 distances in centimetres.

2. [1 mark]What does this program print?

   grid = list(range(64))
   level = grid[16:32]
   print(len(level), level[0], level[-1])

   Answer:

   16 16 31

   grid[16:32] takes items 16 up to but not including 32: sixteen items, the third and fourth rows of eight.

3. [1 mark]Which rows of the depth grid look straight ahead, level with the robot?

   1. ARows 2 and 3
   2. BRows 0 and 1
   3. CRows 6 and 7
   4. DRows 4 and 5
   Answer: A. Rows 2 and 3 are the level rows, items 16 to 31. The bottom rows look down and hit the mat first.

4. [1 mark]One cell of the depth grid reads 400. What does that mean?

   1. ANothing is in range in that direction
   2. BA wall exactly 4 m away
   3. CThe sensor is broken
   4. DSomething is touching the robot
   Answer: A. 400 is the sensor's way of saying it saw nothing. You see it in open space.

5. [1 mark]The camera image is 320 by 240 pixels, and each pixel is three numbers. How many numbers is one frame?

   Answer: 230400. 320 x 240 = 76,800 pixels, times 3 for red, green and blue.

6. [1 mark]A model has only seen the depth grid with the robot facing a wall straight on. The robot turns a little. What happens to the numbers?

   1. AMany of them change, and the model has never seen a view like the new one
   2. BNone change, because the wall has not moved
   3. COnly the middle column changes
   4. DThe model adjusts them automatically
   Answer: A. The numbers depend on exactly where the robot is and which way it faces, so collect data from everywhere the robot will be.

7. [1 mark]What is the name for the small set of numbers you choose to hand to a model, which still tell the situations apart?

   Answer: features. Choosing good features is most of the work. In this module the features are the 16 level readings.

The task: see as numbers

# the two lines every program starts with: the commands, then the robot
from bugbot import *
connect()
# 64 distances, 8 rows of 8
grid = tof_grid()
print(grid)

from bugbot import *
connect()
grid = tof_grid()
level = grid[16:32]
for row in range(8):
    print(' '.join('%3d' % d for d in grid[row * 8: row * 8 + 8]))
print(f'depth: min {min(level)}, max {max(level)}')