วีดีโอ

Back to this after an year! , This probably won't help me with music theory, but def help me overcome my crippling fear of sinusoidal waves 😦

chlape, jsi čech, a ta angličtina Tvoje je brilantní, špička. Dávám to mým slutdentů, co učím anglinu, jako příklad anlginy, které rozumí naprosto každej na planetě, když umí ty slova.

Incredible content, this channel is gold. "Long" technical videos are amazing, the latest trend of short content it's good just for surface explanation or intro to some concepts. I'm a software engineering student and I'm extremely thankful for you publishing your work.

Just calculate traveling salesman problem for every dust particle

There are huege problem with this video. This is'nt AI.... Just a robot with it's program.

Excellent video. People often forget that ML is only a small sub-field within AI. Search algorithms like Minimax and MCTS, A* pathfinding, and even expert systems are all AI. Some are more intelligent and generally applicable than others though.

Great work and explanation of the polygonal models...thanks a lot Vojto 🙂!

what you could do to make the area more efficient is to calculate the area of the work space, use an average amount of points per given area of whatever grid you choose to find the amount of spots that fit in the area, then run a particle simulation where the particles repel each other when they intersect and can only get within its radius of the walls, with no retaining of velocity till they all stop moving, and then you would be left with a pretty even grid that hugs the walls. and then ofc choose the closest neighbors within a given radius

Vojto, thank you for inviting me to colaborate on this video. It has been a true honor for me and I had blast shooting it with you.

(some) Things about the real world that complicate the problem. 1. Sensor limitations. Lidar sensors for example are limited in how far they can see because accuracy degrades with distance. (And there is also a tiny blind spot up close.) 2. Lidar sensor data of obstacles results in a point cloud. If your algorithm uses, clean geometric shapes, they would need to be calculated from the point cloud. 3. Lidar sensor data is in a coordinate system relative to the robot, not global. The robot is not at an angle to the room, things in the room are at an angle to the robot. 4. Not all obstacles are stationary. 5. The robot does not move perfectly straight. Any drift will change how things in the room are angled to the robot. Some form of active correction would need to be used. (The missile knows where it is, because it knows where it isn't.) So yeah, before we can fit a grid to the world and plan out a path, we need to figure out what the world looks like, as everything is vibrating and shaking around us with things popping in and out of existence, doing our best to not get lost. And if we are lucky, maybe we can clean over 90% of the floor.

This looks like it could also make a great algorithm for 3d print layers

Now do stairs. "Agent program"... but you repeat yourself...

D = sqrt(2r²), as the Square can be broken up into 4 equal right triangles with short side of length r

Exactly what I needed, Thanx

I Really like ur videos but could u make a video about how x/y is equal to 0.x repeating except for when x=y and if y is equal to 10^n-1? Id really like to see a Video about that but ima also gonna try myself now

great video but the title is misleading. you can try something like "how do design an autonomous robot from scratch'" or something like that.

Thanks for making this video but nothing covered here was AI or ML.

Clear and good video!

Great visualizations but it's too abstract. Would have loved a more in-depth implementation, preferably with code.

This guy Maths!

I think that probably the most efficient solution would be to instead of "move-turn-move-turn" instead be a series of smooth arcs - so the robot never needs to waste time standing still. My first instinct is also a greedy approach - figure out from any given position which arc could be taken then evaluate which fills the most space - then choose the best and repeat. That will probably be really bad but then you could maybe make it think a couple steps ahead like a chess algorithm. So for every path - and every child path which would be the best. However that would have the issue that naive chess algorithms have which is exploding exponentially more choices evaluated. However unlike chess algorithms it would be much harder to prune obviously terrible choices (such as alpha-beta pruning for chess AI) Regardless - you could apply the techniques that the best of the best chess AIs use: Machine learning. Basicly each choice of arc a ANN will generate a score of how good it thinks that choice will be. Then it picks the best choice and repeats. Idk I could go one with ideas but I just wanna say I love this video: I kinda already know all the topics in this video but I know that me 5 years ago would have loved it so much. Thanks for making it :3

Brother the only thing this channel needs is quick uploads. You literally have the same quality as the best channel.

Amazing video! Would love to learn how you make some of these animations in houdini! I don't think I've seen anyone using it like that before

Interesting way to use fourier transforms. very cool.

10:30 confusing wording, correct wording should be - a vector perpendicular to the face of a polygon.

triangles are always used by GPU to render the object to the screen, but in 3d modeling triangles are not good, you can't loop easily, select a poly ring, do proper edge flow etc. so in modeling people are using quads and simply avoiding edge cases where quads might break, which is not that often tbh, quads are only breaking when a 3d modeler is a beginner and he/she can't produce quad only mesh.

Really cool, very well done!

Amazing video, simple and cool!

This is so well-explained and demonstrated that it feels illegal to watch for free. Thank you Vojtech :)

lovely. I'd love to see more.

Happiness is watching a youtube channel growing from the beginning! All the best❤

Great work as always !

What is the purpose of using the semantic: “the probability of A being visible” instead of “The opacity/alpha value of A”? Just curious.

Great video!

This is gonna blow up, thank you so much for creating this!

This is amazing! I'm shocked your channel isn't bigger, so underrated

bulbublbulb

Vertices edges and faces are not the most simple representation. The edges are redundant. Faces can be represents as triplets of vertices. No need for edges.

A nice overview. Informative.

the best mapgen i've ever seen is this one called 'Terrainbrot' and created by a minetest server owner and developer called Hume2. you can experience it and see a map of the world at 'spawn' on the server 'Tunneler's Abyss' using the free and open source Minetest client and going to the "join game" tab. it has tall realistic mountains, oceans and continents, archipelagos, deep oceans, and is just the most natural and realistic mapgen i've ever seen. by comparison, most other mapgens create bizarre artifacts like mountains cut in half, land tapering outwards as the elevation increases, and all sorts of goofy ugly bits all around like floating bits of land.

Amazing video into this topic. I remember looking into minecrafts procedural generation and perlin noise used to generate the terrains and caves. Wonderful systems at play together.

This was an absolutely brilliant explanation! this whole time I have been wandering blind, changing values without actually understanding what I'm changing. Now I feel like I better understand the concepts behind generating noise, and finally understand what I'm actually doing when tweaking the parameters of fractal noise

Great

After viewing this video, I have some constructive feedback to offer. While the animation and video production quality were impressive, I found that the pacing and structure of the content could use refinement to effectively convey the intended knowledge. It strongly feels like at the beginning you want to talk about true random noise, but at the end you ended up talking about hash function generated noise which is pseudo random and deterministic. You also step by step almost feel like you are talking about generating true random noise procedurally then suddenly introduce another source of randomness required to generate the noise. This is basically luring people in with chocolate and ended up feed them 💩 . The segmentation into 17 sections felt a bit fragmented and the progression from one topic to another could be smoother. For instance, the introduction began with a discussion on true randomness 2:52 and procedural noise generation 3:23 , transitioning into real-world sampling methods like coin flipping, and then delving into molecular motion as another source of noise. Following this, the narrative shifted to sin wave-based noise generation, incorporating an unidentified random source for phase shifts, before introducing Fourier transformation. The revelation that hash functions were being used as the phase shift generator, without a subsequent explanation, left a bit of a gap in understanding. The video's level of technical detail seemed to be caught between a high-level conceptual overview and a low-level technical deep dive. For instance, if the goal was to provide a low-level technical explanation, including actual code and in-depth analysis would have been beneficial. Conversely, if the aim was a high-level overview, the inclusion of complex concepts like Fourier transformation may have been excessive. It might be more effective to focus on core topics like the possibility of true random noise, the generation of pseudo-random noise through computational techniques, and the role of hash functions in ensuring uniform distribution in random generators. A more bottom-up approach, starting from deterministic functions and progressing to more complex noise generation techniques, might provide a clearer, more satisfying learning trajectory. Moreover, the video's objective was a bit unclear-whether it aimed to educate on procedural true random noise generation or realistic terrain noise generation. The content at the beginning and end seemed to address different goals. If the focus is on terrain noise, it would be beneficial to hypothesize on the unique frequency profile of terrain noise and explain the necessity of certain filter combinations to accurately replicate such profiles.

Nice. Please do more! I love experimenting with conplex patterns!

The art of procedural generation is to make something totally random look totally not random.

More. We crave more, mother. Why have you forsaken us, mother? Mother?

This is amazing! I hope you continue to series! ❤❤❤❤❤

Finally I get what those fbm parameters mean🤌 great video! Would love to learn more!

High quality production, great job