Mechatronics & PCB Design

LynxBot

The LynxBot is an Arduino based robot that consists of multiple sensors and a motor controller system that independently controls the speed of each wheel. Working with this robot required accounting for calibration and instabilities for each sensor and actuator. Since these parts are hobby grade, extensive time was invested in developing approaches to offset sensor hysteresis. The robot was designed to be modular and complete a series of tasks, a couple of which are highlighted below.

Navigation by Photoresistor

A photoresistor and Sharp infrared range sensor were used on the LynxBot to accomplish the task of locating and blocking a light. A code was developed to complete the following sequence of tasks: 1) conduct a sweep to identify the direction of the light, 2) move towards the light and stop at the midpoint, 3) conduct another sweep to re-identify the direction of the light, 4) move towards the light and stop within 20 to 30 cm of the light, and 5) pivot 180 degrees and back up slightly to block the light. After significant testing and code refinement, the robot successfully accomplished the task of blocking the light. The initial position of the robot was varied during testing, with initial radius between 2 to 3 m, and across a 30-degree range.

Robot Design

A photoresistor and a Sharp infrared range sensor were added in a forward-facing position to the front of the robot. When the robot initiates its rotational sweep, the photoresistor readings will change depending on the location of the light relative to the robot’s position of heading, allowing for identification of the direction of maximum light exposure. To maintain appropriate distances from the light and the wall behind it, the Sharp sensor is used to actively measure the distance from the robot to obstacles in its path.

To accomplish the goal of blocking the light, a blocker panel was mounted on the back of the robot. As the last step of every test run, the robot pivots and reverses to a position that places the blocker panel near the light, blocking it from view of the initial robot starting position.

The picture below shows an initial starting position of the robot in the test environment. The code is designed such that the LynxBot will be able to execute the procedure starting at any position within a 3 m radius of the light.

Performance Analysis

Code V1

The initial version of the code started by sweeping once and recording highest value from that sweep. It would then reverse the sweep and compare each measured value against the recorded value until it is equal to or greater than it, at which point the sweep would stop. This logic appeared problematic upon testing as the robot would continue the secondary sweep and did not stop. The fundamental flaw in this logic is that it is difficult to guarantee that highest value recorded during the primary sweep will be less than or equal to the secondary sweep. Signal noise plays a critical role in the quality of the recorded values, thus it can exacerbate this issue.

Code V2

The second iteration of code compared the secondary sweep values to 98% of the maximum value recorded to account for the noise in the signal. The photoresistor has a high sensitivity to change in luminance, thus minimal angular direction error occurred when introducing the adjusted comparison value. To further improve the light readings, a lower sweep velocity was employed. This allowed for a greater number of data points per degree of sweep, increasing the probability of identifying a higher reading. These approaches ultimately demonstrated greater success during testing.

Future Areas of Investigation

During a rotational sweep, the robot would shift slightly from its center, increasing the variability of readings over larger sweeps. Thus, to improve performance, the primary sweep angle should be reduced. One approach would be to actively monitor the changes in light readings to detect when the LynxBot is rotating away from the light source.
Multiple times the LynxBot falsely detected that it was close to the wall and would prematurely execute the last light blocking step. This resulted in the blocker being outside of an acceptable distance from light source. One identified cause was that the wiring connections were poor, resulting in a disconnect or higher resistance leading to inaccurate distance readings. To resolve this, stronger wiring connectors should be used going forward.
The photoresistor is highly sensitive to changes in luminance, therefore, a non-linear comparison method should be used during the secondary sweep to account for the proximity to the light source.

Range Sensing with a Sharp IR Sensor

A Sharp infrared range sensor was introduced to the LynxBot to contribute to its vision system. This sensor was calibrated to enable the robot to identify how far it is from a target surface. During the calibration testing, an intermediate program was developed, which used an LED display system to indicate which distance zone the robot was in relative to the target, based on distance thresholds.

Calibration & Testing

The calibration process involved recording sensor values at 5 cm intervals, with both black and white targets being used to account for reflectivity differences in the surfaces. These data points were then used to map a voltage reading to an equivalent distance.

As the sensor approached the target, the white target resulted in slightly higher voltage readings than the black target but did not show any significant differences. This aligns with the SHARP product manual, as reflectivity and color are noted to have minimal effect on the sensor readings.

Additionally, dynamic testing was completed to assess the stability of the system response to changes in position. In the plots below, the voltage is tracked over time, with the sensor starting at 20 cm from the target and moving back and forth by 10 cm approximately every 4 seconds. This procedure was then completed again; however, the sensor starts 40 cm from the target.

Distance Zone Program

The distance zone program uses LEDs to indicate how far away from the LynxBot is from a target. The behavior of the program is described below, with a supporting flowchart outlined on the right.

At the start of the program, a green flashing LED is used to signify that the device is ready for operation. Once the start button is pressed, target range values are set, and the Arduino begins reading and mapping the Sharp voltage to a distance value. The program then checks what range “zone” it is in based on proximity. Upper and lower limit targets will be set depending on zone to prevent flickering. If no previous zone is declared, then no change in target values. Next the Sharp value is compared to the lower target limit, if it is less, then the Yellow LED turns on and the zone is set to 3. If the Sharp value is greater than the upper target limit then Red LED turns on and zone is set to 1. If neither of the previous 2 conditions are true then Green LED turns on and the zone is set to 2. This cycle will repeat until the button is pressed to end the program.

Observations

The non-linearity of the relationship between the millivolt reading of the sensor and the distance results in greater difficulty maintaining a constant distance between the LynxBot and the target at farther distances.
Voltage readings decrease as the distance increases; thus sensor noise will have a greater impact on the measured distance.
Highly dynamic movements reduced evidence of noise as the rate of change of the sensor reading was greater than the error introduced by noise. To address the noise, we can introduce a bypass capacitor (35 microFarad) between the power supply and ground pin, and a low pass filter.

PCB Project

This circuit is designed to use a photoresistor sensor to measure the amount of light in a room and comparator operational amplifier to determine if an LED should be turned on or off accordingly. This system consists of:

10K resistor (x3)
1K resistor (x1)
Photoresistor (x1)
Comparator Operational Amplifier (x1)
LED (x1)
100 uF Capacitor (x1)

An operational amplifier is used to compare the voltages between two voltage divider configurations, one with a 10K resistor and photoresistor, and the other with two 10K resistors. As the amount of light received by the photoresistor decreases, the resistance of it increases, resulting in a higher output voltage from the voltage divider. An opt amp is connected to 9V and GND for its positive and negative voltage terminals. The opt amp has two connection pins, a non-inverting input that connects to the photoresistor based voltage divider, and an inverting pin that connects to the other voltage divider. When the photoresistor receives significantly less light, the voltage at its connected opt amp pin will increase and be higher than the other pin, connecting the 9V to the opt amp output pin, which powers the LED. When the photoresistor is exposed to more light, the voltage at its designated opt amp pin will be less than that of the other voltage divider, therefore the opt amp will connect the output pin to GND and the LED will not light up. A capacitor is also included in the circuit to stabilize fluctuations and spikes in source voltages.

Circuit Schematic

PCB Design

Mechatronics & PCB Design

LynxBot

PCB Project

This circuit is designed to use a photoresistor sensor to measure the amount of light in a room and comparator operational amplifier to determine if an LED should be turned on or off accordingly. This system consists of: ​

10K resistor (x3)

1K resistor (x1)

Photoresistor (x1)

Comparator Operational Amplifier (x1)

LED (x1)

100 uF Capacitor (x1)

Circuit Schematic

PCB Design

This circuit is designed to use a photoresistor sensor to measure the amount of light in a room and comparator operational amplifier to determine if an LED should be turned on or off accordingly. This system consists of: