3.7 Sensor
There are many obstacles to consider when maneuvering in household environments. In order to successfully track a person through this environment we will need a sensor array around the perimeter of the vehicle to detect obstacles and allow the vehicle to make course corrections to avoid them so it can continue to its tracking mission.
Tactile sensors- These are sensors that would require physical contact with objects. Pressure from contact with the object would then be translated by some kind of sensor which would relay the information of an objects presence.
Mechanical Switches- Using metal arms attached to a spring loaded mechanical switches (shown in figure 3.7.1-1). This method is the simplest and lowest tech option. However it is also the slowest and most imprecise method of obstacle detection. The vehicle would need to be far too close to make hard enough contact with the objet to throw the switch, at which point it would need to reverse away from the object before trying a new course correction.
Figure 3.7.1-1 Mechanical switch sensor vehicle diagram
Piezoelectric pressure sensors- Similar to the Mechanical switch sensor system we can use a piezoelectric material in place of the mechanical switch that causes a voltage spike whenever pressure is applied to the material. With this system we could detect objects using less pressure than would be required for the mechanical switch to detect object, however physical contact is still required which again would slow down the procedure of obstacle avoidance as the vehicle would have to reverse out of contact with the object before trying applying a course correction.
Non-tactile- These are sensors that do not require physical contact to detect objects. They typically use a source emitter and a detector for reflections of the source on the environment (Shown in figure 3.7.1-2). Because no physical contact is required to detect objects these sensors could apply course corrections without needing to reverse away from an object, which would cut down on time spent navigating around objects and allow the vehicle to proceed with its mission of tracking and following heat signatures with less delay.
Figure 3.7.1-2 Proximity IR vehicle diagram
IR Sensor-This method would require an IR light array, which would broadcast IR light in the sensing areas. And an IR light sensor, which would detect reflections of the broadcasted light. The intensity of the reflected light is then used to determine proximity with an object as shown in Figure 3.7.1-3 for the VCNL3020 Proximity Sensor w/ Infrared Emitter.
Figure 3.7.1-3 Proximity Value vs. Distance (‘Courtesy of Vishay Intertechnology’)
Sonic- This method uses ultrasonic sound waves to detect objects. Sound is emitted from a source and the reflections of the sound waves off of objects are picked up by receivers. Depending on the location of and number of sensors we can calculate the distance from the object as well as the shape of it (to some degree of accuracy depending on the number of sensors).
|
HC-SR04 (Ultrasonic)
|
VCNL3020 (IR)
|
Operating Range (mm)
|
20-4000
|
1-200
|
Measuring Angle
|
≈ 30°
|
≈ 40°
|
Measurements per sec
|
1 - 16
|
2 - 250
|
Operating Voltage range (V)
|
4.5 - 5.5
|
2.5 – 3.6
|
Operating current (mA)
|
10 - 20
|
10 – 200
|
Price per unit (USD)
|
≈ 1.50
|
≈ 3.20
|
Ultrasonic vs. IR- The two devices we have chosen to compare are the HC-SR04 Ultrasonic Ranging Module, and the VCNL3020 Proximity Sensor w/
Infrared Emitter. Table 3.7.1-1 was made from each item’s data sheet.
Table 3.7.1-1 Comparison of HC-SR04 and VCNL3020
We can see clearly that the Ultrasonic sensor has a much larger operating range compared to the IR sensor, this is a fairly large advantage as it allows us a greater distance to perform course correcting procedures to avoid the obstacle. The measuring angle of the IR sensor is slightly larger; however the difference is not large enough to have much impact on our final decision. The IR has a clear advantage in measurements per sec due to the speed of light being so much greater than the speed of sound, however the extra measurements would just be added processes for the CPU to handle and ultimately unnecessary for our purpose. We also see that the IR system consumes more power, and has a higher cost. However, the biggest downfall of using the IR sensor is that the light it emits encompasses the same spectrum of light that we will be using for the detection of thermal bodies in the tracking system for the vehicle. Using these sensors could cause difficulties with the tracking functionality of the thermal camera by washing out the surrounding area with unwanted IR light.
Share with your friends: |