Lab 9 mapping

Overview

• Goal: Sensing the environment nearby.
• Highlevel:
  1. Rotate the car in place and measure the distance with Time-Of-Flight sensor during the rotation.
  2. Put the orientation and the tof readings in polar coordination, and then convert them into Cartesian coordination.
  3. Repeat the step1&2 in several different location and combine them together to get the map of the room.
• Technical Approach(Pseudocode):

  num_reading=0;
  if (current orientaion is close to set orientaion){
    u=0;
    if (num_reading<num_ori_rotation){
        tof sensing;
        num_reading+=1;
        Setori+=rotate_degree
        calibrate Setori to within (-180,180]
    }else{
        allowrun=false;
    }
  }
  else
    continue on current rotation
  

Data collection and saving

• Point1:
• Point2:
• Point3:
• Point4:
• Then, I used the following python code to save the readings to csv files:
• 
• And read from the saved csv with:
• 
• Next, apply the translation matrix:
• 
• Last, draw it out:
• 

Chanllenges

1. Tunning PID values
   • tunning the PID values has been done in lab6; however, after I used a new battery, the voltage changed from 3.67v to 4.18v, and I have to retune the PID values.
2. Time-Of-Flight sensor 2 doesn’t work
   • Due to unknown reason, the readings from Time-Of-Flight sensor 2 is only the half of that of TOF 1, and it’s not reflecting the room’s shape at all. To resolve this, I decide to discard the readings from TOF2, and get more data from TOF1; This is the reason why I choose read TOF every 10 degree to get 36 readings every cycle instead of the 20 degree in the handout.

Showcase

• Point1:
• 

• 

• Point2:
• 

• 

• Point3:
• 

• 

• Point4:
• 

• 

• Combine all of them together:
• 
• Then draw the lines to connect the dots:
•