Recent Updates Toggle Comment Threads | Keyboard Shortcuts

  • Victor 11:40 pm on June 15, 2017 Permalink | Reply  

    Summer update 

    It’s summer! I had started to post on the quantum club’s Facebook and Twitter pages, and information on the quantum club on this blog should gradually be transferred to the social media pages. Also, the club is developing a website so that all the long writings on the blog can be transferred there. Honestly speaking, I’m not confident on the club’s ultimate purpose – successfully conducting the quantum entanglement experiment described in the paper I’ve been explaining – as I maintain that the experiment is much harder than simply gathering materials and putting them together, so I’m not putting everything into it. However, the club’s leader is a confident Freshman with a lot of brilliant leadership skills for me to learn, and just that is enough for me to do my part for his ambitious plan.

    As for the lawnmower project, I have found a supervisor for the club, and the club should be set up early next semester. Three other sophomores might be willing to work on this project – I knew it would be a matter of time until some sophomores truly interested in programming showed up, since currently all members in Gunn’s computer science club are Juniors, (with the exception of me) and there definitely are programmer Sophomores hidden from my sight.

    I’ve been doing badly on collecting information as always, so I did not have a clue that Gunn’s College and Career Center actually helps students contact start up companies that may need interns, and ultimately I did not find an internship at a company despite my programming skills – all while some friends that barely know algorithms are happily gaining work experience at a start up they contacted through CCC. How sad. Anyways, good for those friends who showed excellent communication traits, and after all I got something as well: I’m currently working for my school as a student assistant, which roughly means updating Chromebooks by the dozens. Definitely not the best internship out there, but I might get to work with some of Gunn’s network equipment if I talk my way in. Still, much learned in the internship-finding process, and I’ll aim for something impressive next semester.

    Meanwhile, on the standardized testing front, I had another bid for a good ACT score, after gaining an impressive 35 without writing last time. For some reason the ACT got harder, and I’m not sure if I can gain 35 on composite again this time. (A classmate of mine confirmed that the questions were indeed harder) To make matters worse, the writing portion did not go particularly well either, and I’m left with regret that I did not prepare for the writing last October when the odds were with me. Still, the test site is only 10 kilometers away, yet the English, Math, and Science preparation is a walk in the park, so a 34 on the test report this time doesn’t translate into total catastrophe either.

     
  • Victor 12:00 am on May 6, 2017 Permalink | Reply  

    Orientation-May 2017 

    April had been a month of rethinking the strategy and plans have been changed accordingly. I will continue to work on the Lawnmower project, albeit in a slower pace and with lower priority, because I realized that the lawnmower project involves more configuration of existing methods – something that high school students aren’t really good at doing compared to companies – than developing entirely new methods with great imagination. Therefore, my extracurricular focus will shift to finding an internship at a lab, possibly working on a “computer creation of art-related designs” project, which is indeed very hard – as you will see in the description below – but involves imagination and developing new methods.

     

    “computer creation of art-related designs” project description:

    Computer Aided Design had been used successfully in media applications for decades, and being able to design on its own is only a matter of time – therefore I aim, with this potentially game changing project, to let the computer take the first steps of designing. Media, especially movies and video games set in the past or the future, require large amounts of object designs that should match the time period the media is set in, which require large amount of human work to complete. Additionally, designers that have accurate knowledge on the historic time period (or a realistic impression of the future, in the case of science fiction) are generally very expensive, yet computers can potentially retrieve information from the time period from the internet in automation. Therefore, it would be extremely gratifying to make software that can design on its own.

    The road to independent designing by computer is undoubtedly a long one, as today’s computers don’t have the level of intuition needed for designing from scratch. Therefore I prepare to start by trying to make the computer modify existing designs to fit certain artistic criteria, through imitating the style of certain human designs. This is surely useful and revolutionary, as movies and video games, along with other types of media, require a colossal amount of art templates, which are very expensive to design. If computers can design a significant portion templates through modifying existing designs according to the style of templates designed by humans intended for use in the same piece of media, a lot of resources can be conserved. For example, a sci-fi movie may require art templates of switches, lamps, fences which can potentially be designed by computers, through imitating the style of human designed templates, enabling a piece of media to include a lot more content on a certain budget.

    How to enable computers to “imitate the style of human designs” is a hard problem to answer. In my opinion, the only realistic way is to understand the style of other designs to be used in the same piece of media and modify a contemporary object according to the style to form a design. For example, if a futuristic-looking switch is needed for a movie, a computer may start by understanding differences between the futuristic furniture in the movie and the furniture today, then modify a switch from today to form a design.

    Understanding differences between complex designs requires a significant amount of intuition, something today’s algorithms cannot possess, as even the generative design algorithms can only optimize immediately measurable factors. However, I believe starting out from designing textures or even color themes for certain objects might not be a very unrealistic plan. After all, texture and color combinations are rather measurable and therefore don’t require too much intuition, yet being able to automatically fill in color combinations and textures is extremely advantageous.

     
  • Victor 12:46 am on April 25, 2017 Permalink | Reply
    Tags:   

    Update 04242017-Lawnmower 

    Still, not a lot of things going on for the Lawnmower project right now…… However, I recently took a deeper look into Google’s machine learning infrastructure, Tensorflow, and found out that it actually has a pre-trained convolutional neural network for image recognition, with over 1000 classes. I am not sure whether “grass” or “lawn” is among the 1000 classes, as it is hard to find this information, but I’m guessing it is. This means that I can potentially use Tensorflow for both Lawnmower part A.1 and B.2. Meanwhile, I am still researching Python, and once the hardware arrives, I will push to make a working prototype before next semester starts.

     
  • Victor 8:45 pm on April 16, 2017 Permalink | Reply  

    Not a lot going on right now 

    Just to fill in the blank, progress is a bit slow, and I’m gathering hardware for the lawnmower prototype, meanwhile preparing for the ACT.

     
  • Victor 12:51 am on April 13, 2017 Permalink | Reply
    Tags:   

    Using google maps to obtain the lawn map 

    So, as we all know, the satellite map from google maps is not exactly high resolution once you zoom to the scale of buildings, so my machine learning program doesn’t really have enough pixels to work with. Therefore, I simply wrote a filter program that makes anything green “lawn”. I must say that the program cannot distinguish between trees and lawns, but hey, at least we can obtain a basic map from google maps. A little step forward. I am working to make a plan to build the working prototype of part A – the robotics team of Gunn High School does not have extra robots to lend me, so I’ll have to build my own in one way or another. I don’t have any teammates to help with this project right now, and I’m kind of in a hard situation. Will storm fry’s and instructables.com for solutions.

    green=lawn
    red=non lawn
    as you can see, trees are recognized as lawns…… But at least we have something to work with.

     
  • Victor 11:41 pm on April 8, 2017 Permalink | Reply
    Tags:   

    google maps API tested-Still room for improvement but very promising 

    As I said before, I plan to use the google maps API to eliminate the need for image stitching and obtain top down images of lawns. Today I read documentation for the API and tested in Python – which I just learned.

    API

    This is the map of Gunn High School, accessed through API. Resolution is not very high (it can be higher as this is not the biggest zoom, but that way the picture is still not very clear), but it is a promising way to obtain photos.python code

    I’m amazed by how simple the code for this is, as python is reputed to be concise.

    The photo of a location can be obtained simply by entering the address into the code.

     
  • Victor 12:45 am on April 8, 2017 Permalink | Reply
    Tags:   

    A bit of detail on the "heralding photons" of the quantum experiment 

    The heralding photons are the first wave of photons to enter the instruments, and there are about 210 heralding photons in total. As the paper explains, the number of heralding photons must be small enough to ensure that the phase broadening of the atomic state is not severe – which means that the heralded state is rather pure – and must be big enough to make the probability of successfully generating the heralded state high enough. With 210 total photons in the heralding photon wave, the probability of detecting the successful generation of a heralded state is 1.5%.

    If you wonder what the “phase broadening of the atomic state” is, what is “rather pure heralded state”, and rather specifically why a large number of heralding photons results in a larger detection probability (although a large number and a large probability does fit together, right?), I very embarrassingly don’t know – I am a member of the quantum physics club of Gunn High School and we’re trying to figure out the paper. I try to make sure everything posted here is correct, and I’ll come back to edit this post when I find out more.

     
  • Victor 3:38 pm on April 7, 2017 Permalink | Reply
    Tags:   

    No more need for image stitching, because…… 

    The original plan for generating the lawnmower map was to use drones to take photos of the lawns and stitch multiple photos together, (this plan was just posted yesterday) but now there is no need. Because, GOOGLE Maps has top down photos of …… the entire US, pre-stitched. So as long as using google maps for the lawnmower is allowed (I’ll check the copyrights), there should be no need for image stitching.

     
  • Victor 4:55 pm on April 6, 2017 Permalink | Reply
    Tags:   

    Next step: making the map 

    As listed in the development plan, part A.2 includes a “map making” part, which generates a map of the lawn (a matrix with each element representing a lawn tile). (The map should of course be to scale). Today I did some research on how this can be accomplished and here are two methods.

    1.Let the lawnmower go along the border of the lawn for a few times, and collect the speed and heading of the lawnmower every 0.1 seconds or so. We then use this data to build a rough, polygon shaped map of the lawn. (With each edge of the polygon the distance the lawnmower traveled in 0.1 seconds.) This map is of course rather rough, but the more laps the lawnmower runs around the lawn, the better the accuracy.

    2.Let a drone fly to the lawn’s location, take a photo of the lawn, and use the lawn recognition program in A.1 on the photo to obtain the map. If the lawn in question is very big, we may have to take multiple photos covering different parts of the lawn, utilize image stitching technology to literally “stitch” the photos together, and then run lawn recognition to get the map. The altitude of the drone can be used to scale the map. (the higher the altitude, the smaller objects on the ground seem to be, right?)

     

    The “image stitching” technology mentioned in method 2 is the technology used to stitch photos from phone cameras to form panorama pictures. I had to do some searching to find this term, as searching “image reconstruction” will only turn up methods used in making 3D images out of multiple 2D ones, which is very useful in situations such as CT scans.

     

    The two methods mentioned above is of course best used in conjunction. More specifically, I plan to generate a precise map with method 2 and then confirm it with method 1, as method 1 produces a rough map that is unlikely to be “wrong” although not very useful, and method 2 may produce an entirely wrong map due to interference from other green plants.

     
  • Victor 12:06 am on March 28, 2017 Permalink | Reply  

    Lawnmower A.3.B specific explanation 

    The A.3.B section determines the optimal direction to mow a section of a lawn in, and I use PCA to do so.

    A straightforward example:

     

    A.2 Example 1

     

    In the picture above you can see that if we mow in the vertical direction, we would have to make many turns and therefore it is very inefficient. However, after displaying the data in a coordinate that has vectors [-0.7,0.7] and [0.7,0.7] as axes, (this is essentially rotating the original picture) we get the following picture, which is much better to mow from vertically. In reality, we simply computer the arctan of primary eigenvector [-0.7,0.7] to get 135 degrees or 315 degrees. This means that the lawnmower can be very efficient if it mows in the 135 degrees or 315 degrees direction (we’ll presume that 0 degrees is north), since doing so is equivalent as mowing in the vertical direction in the second picture.

    A.2 Example 2

    A.2 Example 3

     

     
c
Compose new post
j
Next post/Next comment
k
Previous post/Previous comment
r
Reply
e
Edit
o
Show/Hide comments
t
Go to top
l
Go to login
h
Show/Hide help
shift + esc
Cancel