- Team
Members:
- Paul Potts
- Darrell Rounds
- Michael Sprague
|
|
Return to project paper.
Automated Pet Door April 18, 1997 Paul Potts Darrell Rounds Michael Sprague This has been a fine week for the pet door. We actually have a fully functional prototype! The door itself is constructed of a high-tech polymer which is easy to work with and provided a quick means of prototyping the door mechanism. (Read: Legos) The door has the two way operation desired, with the ability to select one-way operation. It has a sensor to determine if the pet is still in the door before it is closed. This also will reopen the door if something gets in the way while it is closing. The sensor used to monitor the door opening is an infrared beam which is in the plane of the door. This means that the door will never strike the pet. A contact sensor would also be beneficial just to make sure everything is okay, however. In order to get the sensor installed as soon as possible it is just powered by a basic DC voltage. The problem with this is that the ambient lighting has an affect on sensor performance and can cause false alarms. To get around this the IR should be modulated at a frequency of around 10 kHz. This frequency would prevent any cross-talk with the main IR system which is running at 40 kHz. It is yet to be seen if we will have enough time to implement this extra circuitry. Since it is functional as is, we probably won’t bother with it. Something else that could be done is to try to reduce the bulk of the pet’s collar device. The primary thing is to find a suitable battery which is compact and will still give reasonable endurance. One battery that looks promising is the 6 volt Lithium type used in many cameras. It then could be packaged appropriately. We are making good use of the Motorola controller. Our system in its current configuration has four inputs and seven outputs, all simple logic signals (0 or 1). We are looking forward to seeing the other group projects as they get up and running also. It is a good feeling to see a year’s worth of time and effort come to fruition.
April 4, 1997 Paul Potts Darrell Rounds Michael Sprague http://www.egr.msu.edu/~spraguem/ee491.htm This week we have decided to use limit switches to control the motor opening and closing the door. We have set up different LED's to simulate whether the door is opening or closing and which side of the door the pet is on. Also, we have integrated the limit switches to the system to simulate the opening and closing the pet door. Currently, we are implementing simple toggle switches to turn on and off the IR transmitters and the main power. There was one problem with the system, each time the collar transmitted a signal to the door our output LED would turn on for about 5 seconds and cut off. The system was designed to keep the LED on when the RF receiver received. The solution to the problem was simple the collar system did not like the 9 volts so it would not transmit the signal continuously, so we change the voltage to 5 volts which worked out well. There is another problem we are having a little trouble figuring out what kind of motor to use. The motor we have currently draws to much current. If anyone has any suggestions we would gladly consider them.
Automated Pet Door March 28, 1997 Paul Potts Darrell Rounds Michael Sprague This week we have a test program that interfaces the microprocessor to the pet detection system. The test program recieves input from the RF reciever and uses an green led output the signal. We tried to pull a voltage from the processor to power the RF receiver, but each time we tried to test the signal the processor reset itself. The RF reciever was drawing too much current from the microprocessor, therefore it reset itself. Now we have the voltage coming directly from voltage source. At this point we are in the process of incorporating the IR transmitter and reciever into the program. Also now that the first of April is next week, we have to start writing the automatic pet door report and planning the poster board design.
Automated Pet Door March 21, 1997 Paul Potts Darrell Rounds Michael Sprague This week we completed the hardware prototyping for the pet's collar. There are still some bugs with the IR transmitter, however. Although it is working as it is supposed to work minimally, we are having some problems with the IR receiver liking a pulse input rather than a continuous one. We plan to remedy this by using the microprocessor to modulate the IR transmitter, thereby providing the desired input. We still have to interface the RF receiver with the microprocessor. As far as the software is concerned, we plan to use the main program to open and close the door. An interrupt service routine will be used to open the door via a sensor (switch) being activated once the pet is inside the threshold of the door.
Automated Pet Door March 14, 1997 Paul Potts Darrell Rounds Michael Sprague This week we continued to work on our pet door system. We are trying to tweak the IR transmitter and the IR receiver to work at an acceptable range by changing the frequency, but it is not working as we would have hoped to work. So currently we are looking at other factors that we can change to increase the distance. Also, we are going to implement the RF transmitter and receiver to the system. On the software side of things, there are a few bugs to work out in the main program. Next week we plan to progress our system to where the IR receiver turns on the RF transmitter. We also plan to start working on the door implementation and the project report.
Automated Pet Door February 21, 1997 Paul Potts Darrell Rounds Michael Sprague Progress on the Automated Pet Door was very limited in the way of hardware. We are in the process of preparing for the presentations scheduled for this coming Monday. Also, Darrell took the opportunity to speak with Dr. Michael Shanblatt in the Electrical Engineering department concerning burning EEPROM chips. Dr. Shanblatt gave him some good insight as to how to approach the problem, and shared with some various ways of how to approach the problems and to accomlish many of the goals associated with this product. Also we are considering incorporating our EE 485 class, Digital Control and Robotics, into our project as far as the operation of the actual door is concerned. Some suggestions from the class for our system included providing a switch to turn off pet door system if one's gate was open, a way to activate an intruder alarm if the pet door was open, some type of of pet area lighting control at night time.
Automated Pet Door February 14, 1997 Paul Potts Darrell Rounds Michael Sprague This week we were in the laboratory testing our equipment that was purchased. First we tried to reestablish communication from the computer to the M68HC11EVBU board. We familiarized ourselves with the various commands on the pcbug11 software by way of running old EE 331 programs. Next we built the infrared transmitter and receiver circuit. The original circuit diagram was at 8 kHz and the circuit was diagrammed to work at 40 kHz. By increasingt the resistances and capacitances we achieved a reading of a 36 kHz signal. With the increase in frequency, it allowed the infrared receiver to receive a signal from a greater distance. Currently, we are in the process of testing our RF receiver and transmitter. We are using a transistor as a switch to transmit the RF signal to the receiver. In the next week, we plan to continue working on the RF receiver and transmitter circuit and formulating code for the system.
Automated Pet Door February 6, 1997 Paul Potts Darrell Rounds Michael Sprague We are making fine progress on the circuits necessary for the communications and sensing between the pet door and the collar. We have included schematics for the infrared transmitter and receiver. The receiver makes use of a IR detector module which outputs Vcc when it detects infrared light modulated at 40 kHz. The transmitter is simply a IR LED hooked to a 555 timer set to output a 40 kHz signal. The radio transmission and reception is going to be implemented by modifying a commercially available wireless door chime and interfacing it to the other systems. This will require some time probing the existing circuits since the schematics are not available. We decided to use this door chime as it was already in our possession and seemed like the easiest way to get the compact transmitter required for the project. Our next big step is to design of the actual door mechanism and its control. We would like the door to be electrically actuated for opening and closing. This, of course, will be controlled by the microprocessor. The block diagram of the whole system has been included for your reference and review.
Automated Pet Door January 30, 1997 Paul Potts Darrell Rounds Michael Sprague http://www.egr.msu.edu/~spraguem/ee491.htm This week we decided on a final design for the automatic pet door project. We are going to create a system that has an RF and IR transmitter and receiver. The collar will consist of an active IR sensor and a passive RF transmitter. When the animal is within a certain range of the IR transmitter, the IR sensor on the collar will trigger the RF transmitter to send a specific frequency to the RF receiver located on the pet door. The door system will consist of an active IR transmitter and active RF receiver. When the RF receiver receives a signal from the collar the door will open, otherwise the door will stay shut. The system will have a internal counter to regulate the length of time the door will stay open. There will be two independent IR transmitters, one for each side of the door. The microprocessor will alternate their operation. This will enable the system to determine whether the animal is in or outside. In order to eliminate false activation of the pet door, the IR transmitter will emit the IR signal at a certain angle within a certain range. Currently, Michael is researching IR transmitters and receivers that could be used in the project. Also, we are in the process of creating a flow chart for the entire system.
Automated Pet Door January 24, 1997 Paul Potts Darrell Rounds Michael Sprague http://www.egr.msu.edu/~spraguem/ee491.htm Welcome back. This week we were successful in getting our microprocessor to run a simple test program. We still have not settled on a final implementation for the pet detection system. We would like to implement a passive device in the pet collar. However, we still have not been able to determine the feasibility of this system. If you are aware of any systems which use a passive oscillator to trigger a detector then pass the information along to us. The other options still being considered use an active collar. One uses a RF transmitter, the other an IR.
Automated Pet Door December 3, 1996 Paul Potts Darrell Rounds Michael Sprague We did not spend as much time researching this week due to the holiday weekend. However, Michael was helping a friend of his move and brought up the project. He, Al Stockbridge, said that he knew of a theft protection tagging system which made use of small integrated passive oscillators on the individual tags which are attached to the items. The system would send out a signal at the frequency of the oscillator and then listen for an echo which would indicate a tag is in the vicinity. The tags were deactivated by placing them on a pad which induced a current in the circuit sufficient enough to blow a weak link. This system sounds like something that could be used for our pet door. We just don't need the deactivation ability. It would be difficult to have these oscillators at different frequencies and be able to distinguish them. For the system with a transmitting collar it would be best to have all the collars transmit with the same frequency and have a digital code which is used for identification. Have a nice winter break!
Automated Pet Door November 23, 1996 Paul Potts Darrell Rounds Michael Sprague We met with Dr. Bickert of Agricultural Engineering and discussed the Cow ID systems used on dairy farms. He gave us the necessary information to contact several different manufacturers of Cow ID systems. He also told us that there is a farm near Ovid which has one of these systems installed. We are considering taking a field trip there to see what the system looks like and how it operates. He thought that the electronic collars in use by these systems continuously transmit a RF signal which is detected and identified when in close proximity to an antenna. The collar mounted device is approximately the size of a deck of cards. We are currently doing research on the circuits required for transmitting and receiving a 300 MHz signal. We want our collar device to be as compact as possible in order to be wearable by a cat without causing discomfort. We are also considering what different features we want to include on a user control panel and what display device, if any, will be required. We want the user to be able to control the different operating modes of the door as outlined in paragraph two of our progress report dated 10/10/96.
Automated Pet Door November 19, 1996 Paul Potts Darrell Rounds Michael Sprague http://www.egr.msu.edu/~spraguem/ee491.htm We have settled on an implementation strategy for our pet door. We are going to concentrate on developing a radio collar which transmits at a frequency in the area of 300 MHz. The door will have a receiver which will detect this signal at a range of six feet or less. This system is similar to the wireless door chimes which are commercially available with the following exceptions. The door chimes have a range of fifty feet or more and only momentarily transmit the signal. We need to build this transmitter/receiver system and see if it can have reliable detection and reasonable battery life. If this system appears insufficient we will be able to add a touch sensitive pad on the door to be triggered when the pet touches it. Bill Bickert in Biosystems Engineering responded to our inquiry into cow ID systems as follows: Actually, there are several companies marketing systems that are based on cows wearing transponders for identity. -snip- Now, the move is toward using cow ID for sorting purposes--cows programmed for sorting are separated from the rest of the group as they move through a sorting gate that is actuated by a pneumatic cylinder controlled by computer. This gate is usually located somewhere just beyond the exit of the milking parlor. Several commercial dairy farms are using such systems, from at least three different companies. I'd be glad to talk with you about this if you are interested. Bill Bickert Darrell is going to set up a meeting with Bill to see if we can learn more about these "transponders for identity."
Automated Pet Door November 7, 1996 Paul Potts Darrell Rounds Michael Sprague http://www.egr.msu.edu/~spraguem/ee491.htm There exists a technology similar to that which we need to use for pet detection. It is the invisible fencing technology. There is one primary difference, however. That is the fact that in the invisible fence application the pet collar is made aware of its proximity to the boundary, whereas in our system we need the door (boundary) to be aware of the pet's presence. Basically we need to reverse the function of each of the two primary components, the collar and the fixed unit. In researching the invisible fence technology we found a distributor of Radio Fence systems. They replied to an email inquiry with the email address of the actual manufacturer of the product. They have not responded yet. The online information indicated that their systems operate at 300 MHz. Darrell emailed a contact at 3M and he is sending information on an electronic pet door which already exists.
Automated Pet Door October 31, 1996 Paul Potts Darrell Rounds Michael Sprague This week we like the idea of having an active collar which transmits a low power RF tone to be detected by the door when the pet is within close range. We can have a receiver on each side of the door to determine if the pet is inside or outside. Different collars can be set at different frequencies to allow the differentiation of one pet from another. It would be nice to have the collar be in a sleep mode until it is awakened by a signal from the door. This would greatly conserve battery life. This system would be implemented as follows: Collar transmits tone. Receiver bandpasses and envelope detects at this frequency. There would be two of these receiver/detectors, one for each side of the door. The outputs would go to the microprocessor to determine if the door should be opened and which way to open it. The detectors could have threshold components to decide if the pet is in range or the processor could decide this for itself. We need to determine what frequencies should be used for the system. Also, what is needed in the antennae department. What about power consumption, signal to noise ratio, reliability?
Automated Pet Door October 24, 1996 Paul Potts Darrell Rounds Michael Sprague We are currently pursuing the idea of some type of magnetic collar being used to trigger the door. The magnetic system could be implemented with an inductive coil around the perimeter of the door. The inductive properties of the coil would be altered when an external magnetic field is present. This would trigger the unlatching of the door to allow the pet to use it. We need to learn how to achieve this inductive sensing capability. A method for controlling the direction of travel for one-way operation needs to be developed. This could be accomplished using a directional latching device which is controlled by the microprocessor. There are some issues which cannot be addressed with this method. The main one is the differentiation of multiple pets with a collar, this would be necessary if two or more neighbors had the pet door. Some options for this are the varying of the magnetic strength of different collars or an active collar which would transmit a particular code.
Automated Pet Door October 17, 1996 Paul Potts Darrell Rounds Michael Sprague At the last meeting we discussed some ways of designing the pet door. The design that gained the most favor was one that was similar to security systems that used magnetic strips along with electronic gates to detect theft in department stores, music stores, etc. This will require a magnetic tag to be placed on the pet's collar. This would allow only the pet wearing such a collar to come through the pet door. The magnetic tag would trigger a signal to unlatch the door. The pet door would be designed as a single unit to provide easy installation, no more difficult than that of existing pet doors. Now we have to research how the security systems work, i.e. how the sensors detect the magnetic tag. If you know how such systems work, let us know. We are also exploring some alternative detection techniques. These include: optical detection using a reflective collar, a touch sensitive switch to open the pet door, and infrared imaging and identification.
Automated Pet Door October 10, 1996 Paul Potts Darrell Rounds Michael Sprague We, the Associates of PRS Imagineering, are writing you to inform you of the latest happenings of our Motorola design project. As seen in the above statement we've come up with a name for our group: PRS Imagineering Associates. This will be our name for the time being. The product that we would like to implement is called the Automated Pet Door. It is a pet door that will allow only pets that belong to the owners of the house where the pet door is to come in and out using control and sensor technologies. A question to ponder is how to discriminate "my pet from somebody else's pet." This is a question that we intend to answer using the various technologies that are available after considering time and money. Here are some other questions to consider: Is pet in or out? How may times did my pet go out today? Is one able to lock (turn off) the door? Is one way operation available (only let pet in or out). Can a timer be used for operation only during a certain time of day. Is there the possibility of detecting a cat carrying a mouse or any other rodent? What about an LED or other indicator so pet knows when the door is unlatched? As far as the assessment of skills of the group members goes, they are as follows: Paul Potts' expertise falls in the areas of electronic and mechanical hardware design. Darrell Rounds' expertise falls in the areas of assembly language programming and organization. Michael Sprague's expertise falls in the areas of public relations and manufacturing due to the fact that he has access to a machine shop.