Chrome and Nickel plating

I have been looking up chrome plating vs. nickel plating. I was looking at the pros and cons of both and found that chrome plating is cheaper than nickel plating but nickel plating is sometimes stronger then chrome. Nickel-tungsten alloy is about ten times more durable and more environmentally friendly than chrome (here is an article on nickel-tungsten alloy http://web.mit.edu/newsoffice/2009/metal-0519.html). Both of them are very reflective and are almost impossible to tell apart both of them would give us what we need and the price difference is not very large. The problem is that besides nickel-tungsten alloy plating chrome is usually more durable than nickel plating and another problem is that I can't find a place that does that kind of nickel plating most do electroless nickel plating and it is not nearly as strong as chrome plating. I would use nickel-tungsten alloy plating but if we can't find anywhere that does that type of plating than we would have to go with chrome plating.

First Version of the Police ALERT's Outer Shell

Joquin, Sisneros Bros marketing manager, with the sealing container they made for us.

Earlier today I went by Sisneros Brothers to pick up the first version of the outer shell for the Police ALERT. Mr. Edington had visited them a few weeks back and told them what we were looking for. They made this shell for us to begin with. We are going to use it for the time being and design a new one based off what we need. Thanks again to all of Sisneros Brothers for their help!

New Method of Handling Video with RasperryPi and OpenCV

Mine and Albert's Most Recent Progress on the State of the RaspberryPi Vision

In our last post, we talked about our solution to working with video on the RaspberryPI. Since then, we ran into a serious obstacle and had to find a new method.

Our Error

With the original solution, we ran into a sudden error that we could find no solution for. When running our code, the only output was a black window, as apposed to the video stream we had originally received. The only possible explanation we could come up with was a change in our libraries. Instead of re installing, we opted to find a new method.

The New Method

Since the RaspberryPI camera is not a USB device, it is not native to OpenCV; therefore, steps must be taken to "trick" OpenCV into thinking the that the RPI camera is a USB device. (It is beneficial to use the RPI camera because of how little processing power it uses compared to other webcams) Since the tutorial we found on the web was not working, we switched to using another more mainstream tutorial. This tutorial was not our first pick since it involves programming in C, rather than C++. After 6 hours of us sitting at school, we finally made the source code compile and were able to run our program and stream video!

Results

By the time we got things working, we had gained quite a bit more experience with linux, such as with moving files using the command line and other similar things. We got the RaspberryPI to stream video for a specified number of seconds, and then got it to take a picture and store it. 

QueLab Hack Night

Last night Albert, Denton, Zack and I went to QueLab for their Hackers Night. We got a tour of their new facility and then talked to some of their members about our project. Zack and Albert talked to Morgan about developing a better suited tracking algorithm for our project. Denton spent most of the night learning about basic electronics and how to read schematics from Ray. What Denton learned will help him determine the power source we will need to use for the Police ALERT. I talked to Bandit about a variety of things regarding the project. Some of the main points he made was looking into what it would take for our project to be automotive grade, getting in touch with Barbra Stoller a patent attorney who conducts seminars, and collision detection which would allow us to  differentiate between whether a person was approaching the car or  walking by the car. The meeting proved to be very helpful and we hope to attend some of their future Hack Nights.

SODA InvenTeam Goes to Quelab

Zack and Albert getting some advice about OpenCV.

Morgan helping Albert develop a vision algorithm.

InvenTeam all getting great words of advice.

Denton learning about basic electricity.

Math or code joke?

Denton learning how to read a schematic drawing.

Meeting with MMT (Metal Morphosis Technologies)

                     Advice from MMT (Metal Morphosis Technologies)

• Height/length was said to be our biggest problem with the various sizes for each car.
• Dimensions with mirror.
• A Faraday Cage would to block and focus electric fields and to test if signal stops. 
• We need parameters on current and wiring (separate wire and fuse). 
• Keep a notebook up-to-date with sketches, ideas, failures, etc.  
• Solid Works drawings must be perfect (lines connecting).

                  Questions given

• How tall is the model?
• How many AMPS?
• How long does it transmit frequency? 
• Are you interfering with the radio?

                Where do we go from here?

• Get schematics for board
• Blank plates on console for dash board (Havis consoles) 
• Need to add power requirements and find how to maintain a steady level of voltage.
• Find good metal coating
• Quick chip server for searching data base

Students who attended the meeting with MMT:
Chloe Grubb, Taylor Torres, Paige Torres, Denton Shaver, Zack Daniels and Albert Reed

Los Lunas PD Visit to SODA

SODA Students with LLPD Officer Romero look at cruiser.

Students looking for possible Police ALERT mounting spots.

Albert looking for power source and mounting spot for Police ALERT indicator.

Office Romero pointing out the electrical power system for the police equipment.

Zack looking at the dashboard of an LLPD unit.

Researching Police ALERT unit mounting options.

Inside a typical police cruiser.

Meeting with MMT

Jim form Metal Morphosis Technology (MMT) listening to InvenTeam members.

Local Law Enforcement Officers Visit

Today Lt. Owen and Sheriff Deputy Johnson visited the school. We were able to get a good look at both of their cars and the layout of the insides. We also discussed various design options and user interfaces. We realized that the more universal the set up and mounting of the device, the better it will be for patrol work. Johnson gave us a good idea of recording an image every 20-30 seconds in case of a break in. This way we could go back to the general time the break in happened, remove the image distortion, then use the picture to identify the vandal. Lt. Owen also said that having an audio indicator along with the dashboard indicator would help alert them. Owen also asked about have a software on their patrol car's computer that would alert them as well. This way they could be working on a report and monitor motion at the same time. We are going to look into the feasibility of this but it would be a helpful feature. Both officers seemed very interested in the project and enthused to help. Owen was even able to put us in tough with another potential sponsor, Metal Metamorphosis Technologies (MMT), who provide custom modifications to police cars. We will be visiting their office on Friday, December 27th, to discuss the design of the Police ALERT.
We will be sending both officers an online survey for them to forward onto their fellow police officers to ensure a wide range of feedback.

Local law enforcement officers working with InvenTeam on Police ALERT project.

Sheriff's Deputy Johnson with InvenTeam members.

InvenTeam members looking at hardware mounting options for the Police ALERT device.  

Lt. Owen from Bosque Farms PD discussing ideas for the Police ALERT prototype.

Lt. Owen and InvenTeam captain looking at dashboard layout of the police car.

Lt. Owen and InvenTeam members discussing mounting option for the Police ALERT.

Traveling Prototype

Today and yesterday, Maria, Denton, Paige, Taylor, and I have been working on building a prop for presentations. We are using a suitcase to help create a model of a police car to give a visual aid for our audience. Hopefully this object will contribute to creating a lasting memory of our invention and help properly display how our product works and why it is beneficial.

Running Our Own Code on the RaspberryPi

Myself and Albert finally got our own code running on the RaspberryPi. 

Why is this important?

     Up to this point, we had been working on getting OpenCV (our chosen library for computer vision) and the RaspberryPi Camera libraries installed. These are necessary for writing our code for vision detection, as without them we have no programming tools. We finished the installation of OpenCV a few weeks ago, but did not get the raspicam libraries installed until yesterday. The raspicam libraries are used to connect and utilize the RaspberryPi camera. With these things installed, we are now able to write and compile our code. 

How did we do it?

     After getting our libraries installed, we needed to figure out how to write code that would have access to both the OpenCV libraries and the raspicam libraries. This was problematic as both of the libraries were in different locations, and we had no idea how to get them to work together. 

     We solved the issue by locating example programs and storing our program in the same place. Although it is not the most elegant solution, it works. At this point we aren't linux-savvy enough to find a better solution, so we'll stick with this because it works. 

Other notable findings

     One big issue we we're having was figuring out how to compile code that was written. An example would be the program that we wrote that was stored in the sample program workspace, as mentioned above. We had the program written but it was not in an executable form. The fix was to run the  make command within the build directory inside the workspace. This recompiled the code and gave us an executable file to run.

What does the program do?

     The program that we wrote is very simple. All it does it open up a window that streams the view of the camera. The point of it was to make sure everything was functional.

What's next?

     The next step is to begin writing code for motion detection. We also plan to get another RaspberryPi running with opencv, the raspicam library, and python libraries so we can test python against the c++ we have thus far. 

Power Supply/Solar Panels

After doing research on the power requirements, I found that the Raspberry Pi uses about 12v. I also did some research on how to make a battery charger that uses solar panels. For that it looks like we will need a blocking diode to keep the power gathered by the solar panels from escaping back out. I also found a solar panel which can be wired for 12v or 6v use, and will fit on the outer casing of the Police Alert. Since I am not sure about everything that has to go into this I have emailed Que lab in Albuquerque for Advice. I am currently waiting on a reply from them.

A Little Research

A Little Research

Did some research concerning whether or not Police ALERT is a patentable invention and also how we would go about applying for a provisional patent. So far it seems that it would be eligible as long as we met the requirements of The United States Patent and Trademark Office which would probably need to include property interest from a U.S. Government Agency as well as other requirements such as names of all the inventors, their residences, etc..

Leigh's Visit

What I learned during Leigh's visit composed of a powerpoint presentation which spoke in great detail of how Lemselson MIT invenTeams started. She also gave us valuable feedback on our proposal which included mainly adding a larger definition of the problem at hand.

Leighs Visit

During Leigh's visit to SODA I learned more about InvenTeams and feel I have a better understanding of the competition and my job as the Communication Lead. She gave a presentation about Lemelson MIT InvenTeams and what it means to be an inventor along with feedback from the judges and her own personal concerns. This meeting was very helpful to the team.

SODA's InvenTeam receives Recognition from Santa Fe

Leigh's Visit

Leigh's visit to Los Lunas was successful and quite helpful for the team. She gave us a lot of valuable information about our project specifically, as well as knowledge about Lemelson-MIT InvenTeams in general. During a presentation she gave, we learned how qualified professionals and judges thought we could improve our project and what we were doing well. Overall it was a useful and successful visitation.

Yeahhhhhh!

OpenCV is now finally running on the raspberry pi. I finally got it to compile a sample code and run it. I was having trouble linking the libraries for the longest time.

Leigh's Visit

Leigh gave several useful comments to the team. Technical wise, her suggesting the police car give some warning to an approaching intruder was a really good idea. Several of the judges suggested to test each part individually so that we know which part is failing when the prototype is completely assembled.

What I learned

What I learned from Leigh is that we must test our components early and often  and that we need to find a way of telling the threat that the police knows they are there so like if someone gets to close the the light on top of the police car would go off or something like that. 

Leigh's Visit - Feedback and Advise

Leigh visited SODA yesterday.  She covered the basics of the Lemelson-MIT program.  In addition to talking about the LMIT program, she covered some of the feedback and advice provide by the InvenTeam application reviewers.  Leigh also offered the team some advice about moving forward with their project. This included some specific advise about scheduling and meeting with potential sponsors and supporters.

Presenting the Police ALERT Project to Leigh.

Leigh discussing the Lemelson-MIT Program to SODA's team.

Leigh's Visit

Today, Leigh visited the school, she did an overview of InvenTeams and we told her about our project and the progress we've made so far. I was really glad all of the team was able to make it to the meeting! All of the feedback and tips she gave us will be really beneficial. There were a few “hiccups” on our end but I’m sure we will learn from them and be better prepared next time. Looking forward to making some more progress over Christmas break!

Acrylic Cylinders for Allen Displays

Today I have been researching on different methods on mounting a  "weather proof" seal. I Have been talking with Denton who is researching the Solar Panels that would go on top. I am continuing my research to find the best possible way to mount the sea and the proper  Acrylic Cylinder.

Conical mirrors

     I have been researching conical mirrors and what I have fond out so far is that it is best to have a slimmer conical mirror gives you a larger viewing radius than a wider version because, if you multiply the reciprocal of the slope of the mirror by arc-cos it will give you the angle of the viewing radius the larger the slope the lager the viewing radius, though you can have the same viewing radius with both by just changing the the focal length because the smaller the focal length, the larger the aperture or viewing radius will be, but if the focal length is the same than the mirror with the largest slope will give you the largest viewing radius. The only thing is that I have not fond out how to find the viewing radius of the 'teardrop' shaped conical mirror because of the concave shape at the top which makes it very hard for me to preform any math on it because of the concave top, to find the other ones you just have to use basic trigonometry but I don't know how to find the viewing radius of a concave mirror I do have the formula for finding the viewing radius but I do not know how to do it (I don't know if I'm just making this overly complex, but the formula is at the bottom ). I also fond an interesting article of people using conical mirrors with robotics (Here is it URL:http://robotti.wikidot.com/mobilerobotquestions).

Police ALERT - Proof of Concept and soldering

Over the last few days, when InvenTeam students are around, I have had them working on the Police ALERT Proof of Concept test bed.  The base, stand, "mirror", and camera mount is now done.  Albert has been working on redoing the code he had made this summer.  We should have a working model soon.

Seth has been learning how to solder.  He assembled a timing circuit yesterday.  Today, on of the mentors, who is an electronics technician, is going to come in and work with him after school today.

Although he doesn't look it, Seth is thrilled to be soldering a timing circuit together for practice.

Yesterday, I gave Chloe the email to a Sergeant that works for the Los Lunas Police Department. He indicated to send him information regarding the project. He will then have to forward to the Chief of Police. The Chief will have final say on who to operate on the vehicles themselves.

Great Work!

Everyone has been doing some great research for the project, which is very exciting! Getting through the little kinks and details these last couple of weeks has been tedious, but a lot of progress has been made. I'm also very glad we have such great mentors who took the time out yesterday to have a Google hangout with us and help us out. 

Proof of concept test bed

Today I got information from sparkfun about their return policies. We will be returning two CMUcam4 soon. I also worked on building the proof of concept test bed with Maria and Clara. We finished cutting most of the materials and will start assembling them later this week.

Proof of Concept Test Bed

Clara, Maria, and Victoria all working on our proof of concept test bed.

Meeting with Mentors

Everyone working on getting the Police ALERT up and going. Had a successful meeting with three of our mentors, Chana Greene, Wesley Myers, and Tim Hayward.

Chloe, Noah, Albert, Denton, and Zack making progress after getting input from mentors.

SolidWorks and Proof of Concept Test Bed

Today I worked on learning more on how to use Solid Works with Denton, we also went over the current Police ALERT model that Denton had created so that I was up to speed. I also worked on helping build the proof of concept test bed with Denton and Victoria. We cut of some of the materials and painted them black or white depending on the part.

Google Hangout meeting with Tim, Wes, and Chana

Today we had a Google hangout meeting with Tim, Wes, and Chana. Team members present were Chloe, Albert, Noah, Maria, Victoria, Clara, and myself. We worked on figuring out how to install OpenCV to the RaspberryPi, and decided that we are going to continue the project with the RaspberryPi instead of the Arduino Unos. We are now working on compiling our vision code on the RaspberryPi.

Google Hangout meeting with Tim, Wes, and Chana

Today we chatted with our mentors about some of the technical issues that we are having. We are trying to compile opencv onto our raspberry pi. We were informed that we have some corrupt files and are now on the right path. We were also informed that it is best to compile our code straight on the raspberry pi rather than attempting to cross compile it.

Experimenting with Raspberry Pi and Beagle Bone.

After realizing that arduino units are not going to have enough computational power to work effectively in the Police ALERT design, the team started looking into other options.

Chloe, Albert, and Zack tinkering with Pis and Bones.

Today, we worked on getting OpenCV installed on the raspberrypi. It proved to be a rather daunting task. Our plan is to get it running on the raspberrypi so we can see how well it performs.

Today, I worked on figuring out the best way to implement OpenCV code on a microprocessor. We decided that the raspberry pi may be the easiest processor to install OpenCV; therefore, we worked on installing OpenCV on the raspberry pi. I hate a ham sandwich. I enjoyed it.