Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Smart mapping for robots exploring huge areas! 🗺️ This is called Rolling-Bonxai. When a drone or ground robot maps a massive area, say a whole city or massive warehouse, storing the entire map in memory gets expensive fast. You run out of RAM. This solution is simple, only keep...

lukas_m_ziegler's profile picture

Lukas Ziegler

65,176 subscribers

15,711 Aufrufe • vor 4 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

8 Kommentare

Profilbild von Inspector fomo
Inspector fomovor 4 Monaten

Rolling-Bonxai trades full-map RAM for disk I/O latency on reloads. That mirrors the sim-to-real transfer bet I wrote about last month. The real test is whether chunk swaps stay fast enough for continuous locomotion.

Profilbild von Chandra Shekhar 📟
Chandra Shekhar 📟vor 4 Monaten

Where do you keep on finding banger upon banger codes and papers in robotics? Do you taste like the sauce because you are dipped in the sauce? Or is it because you know where to find the best sauce?

Profilbild von Damir Wallener
Damir Wallenervor 4 Monaten

Chunked voxel grid. Watch for… - boundaries -> thrashing -> planning stalls - integration errors -> drift -> can’t find right chunks - map fragmentation -> loop closure -> ghosting - alt/range est. errors -> scale drift -> matching failures It’s simple until it meets the real world. 🙂

Profilbild von MendysRobotics
MendysRoboticsvor 4 Monaten

Sliding-window maps are the pragmatic answer for warehouse-scale AMRs too - you rarely need the full global map in RAM, just local occupancy plus anchors for loop closure. Curious how Rolling-Bonxai handles loop closure corrections once old chunks are already on disk.

Profilbild von Juan Jimeno
Juan Jimenovor 4 Monaten

@facontidavide check this out

Profilbild von Geospatial FM
Geospatial FMvor 4 Monaten

That’s what they’ve always done with computer games

Profilbild von WanderingDuck
WanderingDuckvor 4 Monaten

Your data is too dense. If the space is rectilinear just use a line and width and height.

Profilbild von Johannes Tscharn
Johannes Tscharnvor 4 Monaten

Quite interesting how we developed solutions to these kinds of problems in video games.

Ähnliche Videos

A toilet-cleaning robot from Reflex Robotics. The best part is when the robot dries its hands... That might be the future of robots doing chores for us. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
1:00

A toilet-cleaning robot from Reflex Robotics. The best part is when the robot dries its hands... That might be the future of robots doing chores for us. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

15,766 Aufrufe • vor 7 Monaten

Iceberg calving happens when massive chunks break off a glacier or iceberg, sometimes causing the ice to flip or spin as it crashes into the water. 🌊❄️
0:54

Iceberg calving happens when massive chunks break off a glacier or iceberg, sometimes causing the ice to flip or spin as it crashes into the water. 🌊❄️

AccuWeather

19,514 Aufrufe • vor 7 Monaten

Seat assembling robots! 💺 This robot from KUKA performs an incredibly technology-heavy process. 🦺 It assembles the seats in the car. And it's so hard? Well, when the car moves along with the production line, things get more complicated. The robot also moves on a special additional axis, so it synchronizes with the production line. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:29

Seat assembling robots! 💺 This robot from KUKA performs an incredibly technology-heavy process. 🦺 It assembles the seats in the car. And it's so hard? Well, when the car moves along with the production line, things get more complicated. The robot also moves on a special additional axis, so it synchronizes with the production line. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

10,554 Aufrufe • vor 11 Monaten

Can you deploy a robot in just minutes? Here it navigates a map that was previously created by a phone. The small, sparse points is the phone map, and the thicc creamy points are from the robots own perception. This is the power of Auki's collaborative perception and mapping.
0:43

Can you deploy a robot in just minutes? Here it navigates a map that was previously created by a phone. The small, sparse points is the phone map, and the thicc creamy points are from the robots own perception. This is the power of Auki's collaborative perception and mapping.

Nils

40,760 Aufrufe • vor 1 Monat

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.
0:25

MICROSOFT QUIETLY SHIPPED THE RAG PIPELINE EVERYONE KEEPS TRYING TO BUILD FROM SCRATCH it is called Kernel Memory, and it turns any pile of files into an AI memory that answers with citations. drop your data in and Microsoft's pipeline does the rest: > ingest PDFs, Word, PPT, Excel, web pages, even images > extract the text, chunk it, embed it into vectors > store the vectors with metadata and keep the originals in blob storage > ask a question and it pulls the right chunks with semantic plus hybrid search then it hands the answer back grounded, with links to the exact source and full token usage. and it plugs into whatever you already run: > Azure OpenAI, OpenAI, Ollama, or Anthropic for the models > Qdrant, pgvector, Redis, or Azure AI Search for the vectors > every step is a swappable handler, so you can replace any part with your own one heads up: Microsoft archived the repo in June 2026 and ships it as a reference blueprint, not production. but as a map for how memory and RAG are actually built, nothing is cleaner. run it in one command: docker run kernelmemory/service bookmark this. Microsoft already drew the blueprint, you just build on it.

NO1ennn

85,704 Aufrufe • vor 2 Monaten

Rope climbing robots! 🪢 One objective, multiple approaches to achieve it. Back in 2017 undergraduates from Mechanical Science and Engineering at the University of Illinois utilized many mechanical strategies and tools to create a full menagerie of rope-climbing robots. Amazing creativity and uniqueness of each solution. Which robot is your favorite? Personally, I would say the one at the end dressed as a monkey! 🦍 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
1:15

Rope climbing robots! 🪢 One objective, multiple approaches to achieve it. Back in 2017 undergraduates from Mechanical Science and Engineering at the University of Illinois utilized many mechanical strategies and tools to create a full menagerie of rope-climbing robots. Amazing creativity and uniqueness of each solution. Which robot is your favorite? Personally, I would say the one at the end dressed as a monkey! 🦍 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

13,056 Aufrufe • vor 9 Monaten

Trying a Oner by overlapping keyframes between chunks of video. Julia Harris cinematography. Tony Hou and Jieyi Lee as robots. Jacob Russo as scientist. Bryan Soegondo as the robot Frankenstein.
1:12

Trying a Oner by overlapping keyframes between chunks of video. Julia Harris cinematography. Tony Hou and Jieyi Lee as robots. Jacob Russo as scientist. Bryan Soegondo as the robot Frankenstein.

Jon Finger

17,012 Aufrufe • vor 3 Monaten

Robotics doesn't need to be boring! 😴 Just imagine, you're an only child, you have no one to play football with. Then this fella comes in and you can train your dribbling skills as long as you want (or as long as the battery lasts). Anyway, pretty awesome robot for non-industrial use. Robots can be seen as companions! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:37

Robotics doesn't need to be boring! 😴 Just imagine, you're an only child, you have no one to play football with. Then this fella comes in and you can train your dribbling skills as long as you want (or as long as the battery lasts). Anyway, pretty awesome robot for non-industrial use. Robots can be seen as companions! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

263,015 Aufrufe • vor 10 Monaten

The robot is used to inspect the walls of ships. 🚢 The Magnet Crawler introduced by the University of Bremen is a lightweight and compact climbing robot for magnetic surfaces. The robot has two magnetic wheels driven by geared motors and an elastic tail to improve stability. In the front part of the robot, there is a miniature wireless camera. The robot is remotely teleoperated & the video signal from the camera is displayed to the operator on a portable screen. Let's wait and perhaps the robot will become more autonomous over time. Big fan of robots that make the world a safer place for humans. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:51

The robot is used to inspect the walls of ships. 🚢 The Magnet Crawler introduced by the University of Bremen is a lightweight and compact climbing robot for magnetic surfaces. The robot has two magnetic wheels driven by geared motors and an elastic tail to improve stability. In the front part of the robot, there is a miniature wireless camera. The robot is remotely teleoperated & the video signal from the camera is displayed to the operator on a portable screen. Let's wait and perhaps the robot will become more autonomous over time. Big fan of robots that make the world a safer place for humans. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

21,587 Aufrufe • vor 5 Monaten

Did you know your robot vacuum creates a 3D map of your home? Then they store it on data servers which you have no access to. When you get into the fine print of who really owns the maps… It gets messy. Smart appliances are not a smart thing to have in your home.
1:13

Did you know your robot vacuum creates a 3D map of your home? Then they store it on data servers which you have no access to. When you get into the fine print of who really owns the maps… It gets messy. Smart appliances are not a smart thing to have in your home.

Gubba Homestead

70,582 Aufrufe • vor 10 Monaten

Today's PicoCalc fun: Finding and printing all primes from 0 to 2^16. Because of memory constraints, I had to do it in 15000 integer chunks, but it's pretty fast. The print to console takes longer than the rest of the program. Pretty fun. We'll race a Commodore next.
1:10

Today's PicoCalc fun: Finding and printing all primes from 0 to 2^16. Because of memory constraints, I had to do it in 15000 integer chunks, but it's pretty fast. The print to console takes longer than the rest of the program. Pretty fun. We'll race a Commodore next.

mrdoornbos

13,264 Aufrufe • vor 1 Jahr

Robots can make building tall structures safer. 🧱 The fastening robot from Raise Robotics helps put up walls on high buildings without putting workers in danger. It uses robots to do the risky jobs people used to do. This means fewer accidents and injuries on construction sites. Workers can stay on the ground and control the robot from a safe place. My fear of height just got a fear of height! 🫣 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:31

Robots can make building tall structures safer. 🧱 The fastening robot from Raise Robotics helps put up walls on high buildings without putting workers in danger. It uses robots to do the risky jobs people used to do. This means fewer accidents and injuries on construction sites. Workers can stay on the ground and control the robot from a safe place. My fear of height just got a fear of height! 🫣 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

35,017 Aufrufe • vor 1 Monat

Is Santiago Bernabéu one of the biggest robots?🤯 Its retractable pitch is basically a massive AS/RS system hidden under the field. The engineering numbers are wild: → Six segments, each ~1,500 tons → Perfectly synchronized drives → Millimeter-accurate positioning inside the underground hypogeum How cool is that! Did you see a bigger robot!? ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:24

Is Santiago Bernabéu one of the biggest robots?🤯 Its retractable pitch is basically a massive AS/RS system hidden under the field. The engineering numbers are wild: → Six segments, each ~1,500 tons → Perfectly synchronized drives → Millimeter-accurate positioning inside the underground hypogeum How cool is that! Did you see a bigger robot!? ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

73,827 Aufrufe • vor 10 Monaten

Robotics is entering the mainstream! 👀 Let's accept this fact. Consumptionism – stonks. Possibilities (HW + SW) – stonks. Demographics - not stonks. Modern robots are often designed to interact with our legacy world, whether it's a household, a factory, or a shipyard. That means they often need to work with existing machines, tools, and spaces. In short: retrofitting. Improving old machines with new technologies without major interference in their operation. Modern way of integrating things. You don't have to change the entire environment to deploy a robot. The robot will adapt, just as human would when freshly hired. The promise is to have an army of robots to help with the world designed for us. I also believe that general purpose robots, are not a direct competition for single purpose machines (like a highly efficient specialized packaging machine, for example). It's like adding a robot to this carriage. Specialized machine could be a car ;) ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:59

Robotics is entering the mainstream! 👀 Let's accept this fact. Consumptionism – stonks. Possibilities (HW + SW) – stonks. Demographics - not stonks. Modern robots are often designed to interact with our legacy world, whether it's a household, a factory, or a shipyard. That means they often need to work with existing machines, tools, and spaces. In short: retrofitting. Improving old machines with new technologies without major interference in their operation. Modern way of integrating things. You don't have to change the entire environment to deploy a robot. The robot will adapt, just as human would when freshly hired. The promise is to have an army of robots to help with the world designed for us. I also believe that general purpose robots, are not a direct competition for single purpose machines (like a highly efficient specialized packaging machine, for example). It's like adding a robot to this carriage. Specialized machine could be a car ;) ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

14,396 Aufrufe • vor 8 Monaten

Ben Thompson says the memory makers are Iran, and by making the Strait of Hormuz mistake are inviting the whole industry to build around them "The other thing, the risk they run, is Apple. Apple's lobbying to get Chinese memory, right?" "And what is the number one focus of algorithmic changes? How can we use less memory?" "I think the memory makers probably screw themselves in the long run by creating such a massive target on their back." "I've analogized memory makers to Iran. The issue with the Strait of Hormuz is it's very effective. It's more effective if you don't use it, cuz then it's always hanging out there as something you could do. Now they did it. Turns out it worked." "But the UAE and Saudi Arabia, they're going to build pipelines. They're going to build new ports. They're not going to let this happen again." "It's very painful right now. But say Iran wants to close the Strait of Hormuz in 2035. It's not going to have any effect, because it will have been built around." "My concern for the memory makers is they might have done the same thing. No one's going to let themselves get in this situation again as far as memory goes."
0:58

Ben Thompson says the memory makers are Iran, and by making the Strait of Hormuz mistake are inviting the whole industry to build around them "The other thing, the risk they run, is Apple. Apple's lobbying to get Chinese memory, right?" "And what is the number one focus of algorithmic changes? How can we use less memory?" "I think the memory makers probably screw themselves in the long run by creating such a massive target on their back." "I've analogized memory makers to Iran. The issue with the Strait of Hormuz is it's very effective. It's more effective if you don't use it, cuz then it's always hanging out there as something you could do. Now they did it. Turns out it worked." "But the UAE and Saudi Arabia, they're going to build pipelines. They're going to build new ports. They're not going to let this happen again." "It's very painful right now. But say Iran wants to close the Strait of Hormuz in 2035. It's not going to have any effect, because it will have been built around." "My concern for the memory makers is they might have done the same thing. No one's going to let themselves get in this situation again as far as memory goes."

Fireside Alpha

205,967 Aufrufe • vor 2 Monaten

researchers gave a tiny local model human-style memory and its context limit basically stopped existing a team from MBZUAI, Princeton and Weizmann took a 1B model and rebuilt how it reads. instead of attending to everything at once, the model reads in 1,024 token chunks and passes the important stuff forward through an associative memory, the same way you carry the plot of a book between chapters without rereading them. the design mirrors human memory on purpose. full attention inside a chunk works as short-term memory. the module that carries information between chunks works as long-term memory. they even trained it like a person, starting with short easy texts and raising the difficulty gradually, because memory thrown into the deep end learns nothing. the numbers back it up. the normal model burns 40GB of GPU memory on a long document and collapses hard past its limit, dropping from 0.86 to 0.32 accuracy. the memory version holds 0.71 at double that length while using a flat 12GB no matter how long the input gets. it also needs about 30% fewer FLOPs. the part i keep thinking about is that nobody scaled anything here. they didn't build a bigger model, didn't stretch the window, didn't add compute. they looked at how a brain handles a long day and copied the architecture. a model small enough to run on a consumer gpu now survives documents its own architecture used to choke on. we keep treating intelligence as a compute problem. sometimes it's a memory problem.
0:43

researchers gave a tiny local model human-style memory and its context limit basically stopped existing a team from MBZUAI, Princeton and Weizmann took a 1B model and rebuilt how it reads. instead of attending to everything at once, the model reads in 1,024 token chunks and passes the important stuff forward through an associative memory, the same way you carry the plot of a book between chapters without rereading them. the design mirrors human memory on purpose. full attention inside a chunk works as short-term memory. the module that carries information between chunks works as long-term memory. they even trained it like a person, starting with short easy texts and raising the difficulty gradually, because memory thrown into the deep end learns nothing. the numbers back it up. the normal model burns 40GB of GPU memory on a long document and collapses hard past its limit, dropping from 0.86 to 0.32 accuracy. the memory version holds 0.71 at double that length while using a flat 12GB no matter how long the input gets. it also needs about 30% fewer FLOPs. the part i keep thinking about is that nobody scaled anything here. they didn't build a bigger model, didn't stretch the window, didn't add compute. they looked at how a brain handles a long day and copied the architecture. a model small enough to run on a consumer gpu now survives documents its own architecture used to choke on. we keep treating intelligence as a compute problem. sometimes it's a memory problem.

Alex Veremeyenko

16,147 Aufrufe • vor 3 Monaten

For my video on Fixing voyager 1 I wanted to try an illustrate the code relocation, so I made this little video showing the chunks of code getting relocated out of the bad memory into the free space they found.
0:42

For my video on Fixing voyager 1 I wanted to try an illustrate the code relocation, so I made this little video showing the chunks of code getting relocated out of the bad memory into the free space they found.

Scott Manley

27,832 Aufrufe • vor 1 Jahr

I spy with my little eye! 👁️ The BBC came up with a wheeled robot disguised as elephant dung to film wildlife up close. The robot rolled alongside a group of elephants, allowing it to roll quietly alongside the animals without spooking them. It's 2026 and we have bots pretending to be elephant poop to shoot content! 💩 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:58

I spy with my little eye! 👁️ The BBC came up with a wheeled robot disguised as elephant dung to film wildlife up close. The robot rolled alongside a group of elephants, allowing it to roll quietly alongside the animals without spooking them. It's 2026 and we have bots pretending to be elephant poop to shoot content! 💩 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

17,808 Aufrufe • vor 3 Monaten

Here it is: (almost) everything from the Starfield Chunks menu!
0:25

Here it is: (almost) everything from the Starfield Chunks menu!

KatOfWhimsy

1,047,231 Aufrufe • vor 3 Jahren

A pentagon robot! 🖐🏽 Compared to typical SCARA models, it delivers higher speed, improved stability, better space efficiency, and can operate smoothly in tight areas, even through singularity positions. These robots are a good fit for handling complex tasks in challenging environments where conventional robots often fall short. Have you ever seen a robot like this? 👀 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:24

A pentagon robot! 🖐🏽 Compared to typical SCARA models, it delivers higher speed, improved stability, better space efficiency, and can operate smoothly in tight areas, even through singularity positions. These robots are a good fit for handling complex tasks in challenging environments where conventional robots often fall short. Have you ever seen a robot like this? 👀 ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

42,565 Aufrufe • vor 7 Monaten