For the following three questions (1a-1c) refer to the first figure
- 1a. In a one dimensional environment what does the convolution mask [1, -1] do? (This convolution mask is different from the one in class. We know this, but if you understand the process of convolution this will not be a problem.)
- 1b. If this mask were expanded to [1, 1, 1, -1, -1, -1], how would the effect be different?
- 1c. If you wished to detect
only the indicated edge, how would you post process the output of the
convolution of the original signal with the [1, 1, 1, -1, -1, -1]
mask? A qualitative sketch of the output of this convolution
operation will be helpful.
Figure 1 - 2. Convolve the follwing image with the given filters. Plot the results. You may deal with the
edges as you wish (padding with zeroes, using nearest neighbor, and cropping are all valid). If you choose
not to use a computer to produce plots, please be as accurate and neat as possible.
[ 5 | 5 | 5 | 9 | 9 | 9 | 8 | 7 | 6 | 5 | -5 | 0 | 0 | 0 | 0 ]
- (a) [ -1 | 1 ]
- (b) [ 1 | -1 ]
- (c) [ 1 | 0 | 0 | 1 ]
- (d) a filter of your choice.
- (a) [ -1 | 1 ]
- 3. We suggest that you print out the sample environment
and draw your answer on the printout.
- Draw the resulting path from the wavefront planner for the mobile robot using the L1 metric. Remember to take into account the configuration space of the robot
- Is the planner outlined below complete? Prove your answer.
- 1.) Choose a global direction left or right. (Without loss of generalitly, the planner will be described with right.)
- 2.) Move towards the goal in a straight line.
- 3.) If the robot hits an obstacle, declare the point hi to be the point where the robot hit the obstacle.
- 4.) Turn right and circumvent the obstacle until:
- a.) The robot returns to hi without defining an li: return(no path)
- b.) The robot reaches a point where it can continue in a straight line towards the goal: define the point li, go to step 2