/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
https://www.innovationportal.org/portfolio/14938
SENIOR DESIGN PROJECT
Moderators: kgudger, bfinio, MadelineB, Moderators
-
deleted-280627
- Posts: 8
- Joined: Tue Apr 21, 2015 6:00 am
- Occupation: Student
- Project Question: Creating a camera mount and system that would use Passive infrared sensors/servo motor at certain angles to track the motion of a person.
- Project Due Date: may 28
- Project Status: I am conducting my experiment
SENIOR DESIGN PROJECT
I'm in EDD (Engineering Design and Development) and for part of the project our work needs to be reviewed by an "expert". My project is an auto-tracking camera mount. The mount will be able to hold many different types of cameras that most households own. Once on the mount I am using PIR sensors at different angles to detect a person. Thanks
https://www.innovationportal.org/portfolio/14938
https://www.innovationportal.org/portfolio/14938
-
deleted-249560
- Posts: 496
- Joined: Thu Nov 20, 2014 1:35 pm
- Occupation: Science Buddies content developer
- Project Question: N/A
- Project Due Date: N/A
- Project Status: Not applicable
Re: SENIOR DESIGN PROJECT
I created an account but it won't let me into that portfolio with the link you provided. You may need to see if there's some permission you have to set to make it viewable.
PIR sensors generally have a wide range so you'll need to be creative about the direction you point them and how you interpret the data. How do you plan to handle the case with multiple people in a room? That could be a challenge as well. On the mount, consider how much a camera might weigh and choose your servos accordingly. A standard cheapo hobbyist servo might ave trouble managing the tilt on a larger camcorder unless you get it really well balanced.
It's a fun project and I'd look forward to seeing your page if you can figure out how to make that possible.
Howard
PIR sensors generally have a wide range so you'll need to be creative about the direction you point them and how you interpret the data. How do you plan to handle the case with multiple people in a room? That could be a challenge as well. On the mount, consider how much a camera might weigh and choose your servos accordingly. A standard cheapo hobbyist servo might ave trouble managing the tilt on a larger camcorder unless you get it really well balanced.
It's a fun project and I'd look forward to seeing your page if you can figure out how to make that possible.
Howard
-
deleted-280627
- Posts: 8
- Joined: Tue Apr 21, 2015 6:00 am
- Occupation: Student
- Project Question: Creating a camera mount and system that would use Passive infrared sensors/servo motor at certain angles to track the motion of a person.
- Project Due Date: may 28
- Project Status: I am conducting my experiment
Re: SENIOR DESIGN PROJECT
I think I need to invite you to see it via email, whether you are able to give it out on this website I'm not really sure.
-
deleted-249560
- Posts: 496
- Joined: Thu Nov 20, 2014 1:35 pm
- Occupation: Science Buddies content developer
- Project Question: N/A
- Project Due Date: N/A
- Project Status: Not applicable
Re: SENIOR DESIGN PROJECT
Yes, that's a hard and fast rule on this site - no email exchange and no personal information. You can email me at my sciencebuddies.org staff account as 'howard' (I'm specifically not putting the address together to avoid having spammer robots pick it up and flood my mailbox). Please don't send any more than the link I'll need to look at your project. No personal contact information please.
Howard
Howard
-
deleted-280627
- Posts: 8
- Joined: Tue Apr 21, 2015 6:00 am
- Occupation: Student
- Project Question: Creating a camera mount and system that would use Passive infrared sensors/servo motor at certain angles to track the motion of a person.
- Project Due Date: may 28
- Project Status: I am conducting my experiment
Re: SENIOR DESIGN PROJECT
I tried to invite you, I don't know if I got the email quite right. But I hope you got the invite!
-
deleted-249560
- Posts: 496
- Joined: Thu Nov 20, 2014 1:35 pm
- Occupation: Science Buddies content developer
- Project Question: N/A
- Project Due Date: N/A
- Project Status: Not applicable
Re: SENIOR DESIGN PROJECT
You did - I'm sorry I couldn't just include it but you know how the spammers are. Once they get your name, your mailbox is lost.
If I can figure out the horrible interface on that site I'll comment there.
Howard
If I can figure out the horrible interface on that site I'll comment there.
Howard
-
deleted-280627
- Posts: 8
- Joined: Tue Apr 21, 2015 6:00 am
- Occupation: Student
- Project Question: Creating a camera mount and system that would use Passive infrared sensors/servo motor at certain angles to track the motion of a person.
- Project Due Date: may 28
- Project Status: I am conducting my experiment
Re: SENIOR DESIGN PROJECT
It said that you started a review but I couldn't see if you posted anything. If you want to you could just post your thoughts here. But yes everyone agrees that the site isn't very good.
-
deleted-249560
- Posts: 496
- Joined: Thu Nov 20, 2014 1:35 pm
- Occupation: Science Buddies content developer
- Project Question: N/A
- Project Due Date: N/A
- Project Status: Not applicable
Re: SENIOR DESIGN PROJECT
It says there is one complete and one active. I don't know why. This is a copy/paste of the summary.
Please let me know if you have any questions about what i said. I didn't grade the sections or worry about the writing style quality (which was good, BTW).
Howard
A Presentation and justification of the problem
?
This is fine as a summary. The statistics certainly speak to the need for a better way of getting evidence.
B Documentation and analysis of prior solution attempts
?
I'd suggest you include a survey of some of the commercial tracking cameras available. There seem to be quite a few, like this one: http://www.ezwatch.com/ez-track-dome.html (whose site you refer to elsewhere) that are using the imagery from the video to determine motion and identify objects. As you noted, these are very expensive but it would be worth discussing how they work and what the differences are between these are your design.
C Presentation and justification of solution design requirements
?
This shows good thought into the issues and i like that one of your goals is to make it inexpensive. Whether you hit that price point or not (I don't expect anywhere near the $40 mark) it should be far cheaper than the $1000+ that the commercial solutions seem to offer.
D Design concept generation, analysis, and selection
?
There are no comments for this element review.
E Application of STEM principles and practices
?
Smart to turn the sensors sideways! That does give you a more precise positioning system at the expense of adding perhaps an extra sensor. That's still a good tradeoff if it helps catch a burglar.
Your explanation of the science and tech here is both sound and easy for others to understand.
F Consideration of design viability
?
You do seem to have done your research on parts & availability. You'll find that the direct purchase from China is rather difficult so I'd not count on the $20 version - at least not right away.
G Construction of a testable prototype
?
The sensor arrangement and support/baffle system looks fine. I can't say whether it will work or not looking at it, but that's what your testing will confirm. It's certainly a promising design.
I'm concerned about your camera mount. If the center of gravity of the camera is not directly over the servo shaft, you'll be putting a torque on the shaft which will make it harder for the shaft to turn, reduce the load capacity and accelerate the wear on the servo. That's why I had suggested a lazy susan sort of system but yours might be helped by simply adding some small support wheels holding up the edges of the platform. Three or four small casters pushing up on the bottom of the platform, arranged 120 or 90 degrees apart would let the camera weight be borne by the casters while the servo only had to handle the rotation.
The code also seems to identify the sensor area with new motion and then snap the camera to that sector. Ignoring the obvious - that servos tend to make noise which will attract the thief's attention - the camera motion will be jerky and make the video harder to follow. You may also run into a situation where the sensors cause the code to rapidly switch back and forth between two sectors. Consider changing your algorithm so that if one sensor is now more interesting than the one to one side of it, move the servo slowly (perhaps incrementally changing from one position to the other) and think about some algorithm which prevents it from just oscillating back and forth. Conceptually what you've done looks good but I'd be curious if the video is hard to watch as the camera snaps from position to position.
Do think about ways of muffling the noise. Can you add some foam or insulation in the area beneath the platform to help muffle the servo noises?
H Prototype testing and data collection plan
?
A good plan. Please consider the previous suggestion of adding supports around the servo with casters so that the camera weight isn't twisting the servo shaft. Also, what happens if multiple bad guys start walking around the room at the same time?
I Testing, data collection and analysis
?
There are no comments for this element review.
J Documentation of external evaluation
?
There are no comments for this element review.
K Reflection on the design project
?
There are no comments for this element review.
L Presentation of designer's recommendations
?
There are no comments for this element review.
M Presentation of the project portfolio
?
There are no comments for this element review.
N Writing like an Engineer
?
There are no comments for this element review.
Please let me know if you have any questions about what i said. I didn't grade the sections or worry about the writing style quality (which was good, BTW).
Howard
A Presentation and justification of the problem
?
This is fine as a summary. The statistics certainly speak to the need for a better way of getting evidence.
B Documentation and analysis of prior solution attempts
?
I'd suggest you include a survey of some of the commercial tracking cameras available. There seem to be quite a few, like this one: http://www.ezwatch.com/ez-track-dome.html (whose site you refer to elsewhere) that are using the imagery from the video to determine motion and identify objects. As you noted, these are very expensive but it would be worth discussing how they work and what the differences are between these are your design.
C Presentation and justification of solution design requirements
?
This shows good thought into the issues and i like that one of your goals is to make it inexpensive. Whether you hit that price point or not (I don't expect anywhere near the $40 mark) it should be far cheaper than the $1000+ that the commercial solutions seem to offer.
D Design concept generation, analysis, and selection
?
There are no comments for this element review.
E Application of STEM principles and practices
?
Smart to turn the sensors sideways! That does give you a more precise positioning system at the expense of adding perhaps an extra sensor. That's still a good tradeoff if it helps catch a burglar.
Your explanation of the science and tech here is both sound and easy for others to understand.
F Consideration of design viability
?
You do seem to have done your research on parts & availability. You'll find that the direct purchase from China is rather difficult so I'd not count on the $20 version - at least not right away.
G Construction of a testable prototype
?
The sensor arrangement and support/baffle system looks fine. I can't say whether it will work or not looking at it, but that's what your testing will confirm. It's certainly a promising design.
I'm concerned about your camera mount. If the center of gravity of the camera is not directly over the servo shaft, you'll be putting a torque on the shaft which will make it harder for the shaft to turn, reduce the load capacity and accelerate the wear on the servo. That's why I had suggested a lazy susan sort of system but yours might be helped by simply adding some small support wheels holding up the edges of the platform. Three or four small casters pushing up on the bottom of the platform, arranged 120 or 90 degrees apart would let the camera weight be borne by the casters while the servo only had to handle the rotation.
The code also seems to identify the sensor area with new motion and then snap the camera to that sector. Ignoring the obvious - that servos tend to make noise which will attract the thief's attention - the camera motion will be jerky and make the video harder to follow. You may also run into a situation where the sensors cause the code to rapidly switch back and forth between two sectors. Consider changing your algorithm so that if one sensor is now more interesting than the one to one side of it, move the servo slowly (perhaps incrementally changing from one position to the other) and think about some algorithm which prevents it from just oscillating back and forth. Conceptually what you've done looks good but I'd be curious if the video is hard to watch as the camera snaps from position to position.
Do think about ways of muffling the noise. Can you add some foam or insulation in the area beneath the platform to help muffle the servo noises?
H Prototype testing and data collection plan
?
A good plan. Please consider the previous suggestion of adding supports around the servo with casters so that the camera weight isn't twisting the servo shaft. Also, what happens if multiple bad guys start walking around the room at the same time?
I Testing, data collection and analysis
?
There are no comments for this element review.
J Documentation of external evaluation
?
There are no comments for this element review.
K Reflection on the design project
?
There are no comments for this element review.
L Presentation of designer's recommendations
?
There are no comments for this element review.
M Presentation of the project portfolio
?
There are no comments for this element review.
N Writing like an Engineer
?
There are no comments for this element review.
-
deleted-280627
- Posts: 8
- Joined: Tue Apr 21, 2015 6:00 am
- Occupation: Student
- Project Question: Creating a camera mount and system that would use Passive infrared sensors/servo motor at certain angles to track the motion of a person.
- Project Due Date: may 28
- Project Status: I am conducting my experiment
Re: SENIOR DESIGN PROJECT
I agree with your suggestions that the servo wouldn't balance very well, that it would also be noisy tipping of a burglar, etc. I will try to fix those problems. Thanks so much for the help, I appreciate it!