INTRODUCTION
The purpose of this project was to take a traditional remote controlled car and create our own control mechanisms. To control the car, we used a dual-axis accelerometer and LEDs (light emitting diodes) configured as photo-detectors. The control mechanism was selected using a single pole dual throw (spdt) switch was wired in a single pole single throw (spst) configuration. The two-axis accelerometer was used as a tilt sensor that detected how far the chip was angled from its neutral, flat position. An analog voltage was produced for both the x- and y-axes that corresponded to how far the accelerometer was tilted, and in what direction.
A custom LED array was also used to control the car. Though LEDs are traditionally thought of as light producing elements, they also produce a voltage corresponding to how much light they receive – brighter environments produce a greater voltage across the LED. We took advantage of this fact by causing the decrease in voltage by blocking light from entering one of the four LEDs (forward, reverse, left and right) to cause the car to move in the appropriate direction (See accelerometer & led sensor array unit).
To implement a safer car, an infrared distance sensor was used to continually scan the area in front of the car. The sensor produced an analog output voltage proportional to how close an object was. When an object was detected, the car momentarily reverses to avoid a collision.
PROJECT OVERVIEW
We chose this project because of our love of cars and the hardware aspect of gadgets. We thought that it would be a novel idea to be able to control the car with one hand by using the dual-axis accelerometer. To provide a contrast to this one-handed approach, we also decided to implement a spin on the traditional two-handed control of the car by using LEDs as photo-detectors. While the use of LEDs would still require two hands, the cost would be even less than the two joystick control scheme that the car came with. We wanted to implement both control schemes, then compare and contrast them at the end.
The car was able to operate reliably. We found that even when the transmitter and receiver pair was close to each other, packets were dropped. This became a problem in a project such as ours because fast reactions were necessary. The controller must operate the car in a very deliberate manner to produce the right reactions from the car. Because there was some lag between the controller and the car, the operator must be aware of how to operate the car using this out control mechanism.
The IR sensor that was attached to the front of the car had a fairly small side-to-side range of detection. This meant that incoming objects must be close to the centerline of the car to be detected. When operating the car in real world conditions, this means that not all collisions can be avoided. This could have been remedied by having two additional sensors at both sides of the car, but that would have caused us drastically go over our $50 budget.
The accelerometer was more reliable method to control the car than LEDs. The LEDs were heavily dependant on the amount of ambient light that was available, which can vary widely. The amount of voltage that was produced by the LEDs varied between different colors. This would mean that any mass produced system using LEDs in this manner would have to be individually calibrated, which would negate the low costs of using LEDs. Eliminating LED controls would have reduced the size of the transmitter unit by over 50%.
We have conducted an endurance test with extra batteries and found out that the car could operate for 30 to 45 minutes of continued use. We believed this was on par with the original specifications of the car. Instead of six AA batteries, the car only required two 9V batteries. This resulted in the car operating faster than it originally was, because we are supplying approximately 7.5V to the motors. The effects were dramatic whenever the batteries are exhausted, as the car slowed to a crawl. The transmitter unit optimized power most effectively and was able to sustain the use of its battery better than the car itself.
We were able to keep the car’s design unaltered, except for the collision sensor attached to the front bumper. This was one of the goals for this project. The h-bridge circuit occupied most of the car’s internal circuitry. Had we been able to use an h-bridge chip, the size of the internal circuitry within the car would have drastically been reduced.
RF interference was a concern throughout the testing stages of this project because there were a few other teams using the same transmitter/receiver pair that operated at the same frequency as ours. In order to properly test the functionality of our car we the ECE lounge to eliminate the possibility of interference. Testing the car in the ECE lounge allowed us to very that the range of operation between the car and controller was 30 feet.
CONCLUSION
We are extremely satisfied with how the project turned out. We were able to successfully implement both accelerometer and LED sensor controls. As expected, the accelerometer was a better control mechanism than the LEDs. To improve the LED controls in the future, we could possibly have another LED to be used solely to calibrate the ambient light threshold wherever the car is being operated. This would adjust to changing light conditions for a more robust control system.
The Mega32 chip was more powerful than we needed. Because the speed and processing power required for this project was not stringent, we used approximately 55% of the chip’s functionalities.
We made use of the Wireless Protocol by Meghan Desai. This greatly simplified the transfer/receiver procedure. However, we had to correct some errors in the driver file before it would operate. We also made use of the receiver box to make sure that we had wired up the transmitter correctly.
Link: http://instruct1.cit.cornell.edu/courses/e...l%20Project.htm
The purpose of this project was to take a traditional remote controlled car and create our own control mechanisms. To control the car, we used a dual-axis accelerometer and LEDs (light emitting diodes) configured as photo-detectors. The control mechanism was selected using a single pole dual throw (spdt) switch was wired in a single pole single throw (spst) configuration. The two-axis accelerometer was used as a tilt sensor that detected how far the chip was angled from its neutral, flat position. An analog voltage was produced for both the x- and y-axes that corresponded to how far the accelerometer was tilted, and in what direction.
A custom LED array was also used to control the car. Though LEDs are traditionally thought of as light producing elements, they also produce a voltage corresponding to how much light they receive – brighter environments produce a greater voltage across the LED. We took advantage of this fact by causing the decrease in voltage by blocking light from entering one of the four LEDs (forward, reverse, left and right) to cause the car to move in the appropriate direction (See accelerometer & led sensor array unit).
To implement a safer car, an infrared distance sensor was used to continually scan the area in front of the car. The sensor produced an analog output voltage proportional to how close an object was. When an object was detected, the car momentarily reverses to avoid a collision.
PROJECT OVERVIEW
We chose this project because of our love of cars and the hardware aspect of gadgets. We thought that it would be a novel idea to be able to control the car with one hand by using the dual-axis accelerometer. To provide a contrast to this one-handed approach, we also decided to implement a spin on the traditional two-handed control of the car by using LEDs as photo-detectors. While the use of LEDs would still require two hands, the cost would be even less than the two joystick control scheme that the car came with. We wanted to implement both control schemes, then compare and contrast them at the end.
The car was able to operate reliably. We found that even when the transmitter and receiver pair was close to each other, packets were dropped. This became a problem in a project such as ours because fast reactions were necessary. The controller must operate the car in a very deliberate manner to produce the right reactions from the car. Because there was some lag between the controller and the car, the operator must be aware of how to operate the car using this out control mechanism.
The IR sensor that was attached to the front of the car had a fairly small side-to-side range of detection. This meant that incoming objects must be close to the centerline of the car to be detected. When operating the car in real world conditions, this means that not all collisions can be avoided. This could have been remedied by having two additional sensors at both sides of the car, but that would have caused us drastically go over our $50 budget.
The accelerometer was more reliable method to control the car than LEDs. The LEDs were heavily dependant on the amount of ambient light that was available, which can vary widely. The amount of voltage that was produced by the LEDs varied between different colors. This would mean that any mass produced system using LEDs in this manner would have to be individually calibrated, which would negate the low costs of using LEDs. Eliminating LED controls would have reduced the size of the transmitter unit by over 50%.
We have conducted an endurance test with extra batteries and found out that the car could operate for 30 to 45 minutes of continued use. We believed this was on par with the original specifications of the car. Instead of six AA batteries, the car only required two 9V batteries. This resulted in the car operating faster than it originally was, because we are supplying approximately 7.5V to the motors. The effects were dramatic whenever the batteries are exhausted, as the car slowed to a crawl. The transmitter unit optimized power most effectively and was able to sustain the use of its battery better than the car itself.
We were able to keep the car’s design unaltered, except for the collision sensor attached to the front bumper. This was one of the goals for this project. The h-bridge circuit occupied most of the car’s internal circuitry. Had we been able to use an h-bridge chip, the size of the internal circuitry within the car would have drastically been reduced.
RF interference was a concern throughout the testing stages of this project because there were a few other teams using the same transmitter/receiver pair that operated at the same frequency as ours. In order to properly test the functionality of our car we the ECE lounge to eliminate the possibility of interference. Testing the car in the ECE lounge allowed us to very that the range of operation between the car and controller was 30 feet.
CONCLUSION
We are extremely satisfied with how the project turned out. We were able to successfully implement both accelerometer and LED sensor controls. As expected, the accelerometer was a better control mechanism than the LEDs. To improve the LED controls in the future, we could possibly have another LED to be used solely to calibrate the ambient light threshold wherever the car is being operated. This would adjust to changing light conditions for a more robust control system.
The Mega32 chip was more powerful than we needed. Because the speed and processing power required for this project was not stringent, we used approximately 55% of the chip’s functionalities.
We made use of the Wireless Protocol by Meghan Desai. This greatly simplified the transfer/receiver procedure. However, we had to correct some errors in the driver file before it would operate. We also made use of the receiver box to make sure that we had wired up the transmitter correctly.
Link: http://instruct1.cit.cornell.edu/courses/e...l%20Project.htm