my controller measured a disturbance I hid from it and I am so proud of myself a week ago I did not know what PID was but in a Week I made this yay

Vega is ship ready I recorded this video to seeits angle limits! https://www.youtube.com/watch?v=oK2FTemJwAE Hope I can get funding soon

Update on Bionix I am finished with my basic hand structure i printed it irl and saw some errors and then fixed it in CAD next is to work on the thread routing channels!! keep following and checking out my github

Vega Is ready to ship!! check it out at github.com/Araadh3111/vega

I added ai usage in readme!

Hi, So I have been working as u can see from logged hours, MY GIMBAL IS DONE HOORAYY!!!!!!!!!!!!! I am so so excited! I spent so much time on it, for the last hours I made two control horns two servo mounts attatched it all fixed height issue made a BASE and it was so fun here are some pics to show u what im talking about:-

Vega Devlog 4! I am done with designing the gimbal system with two rotational axis and next time will work on the base for this time I made inner cylinder with two tabs sticking out, its also the part that contains the rocket motor its diameter is 24.250 mm and the wall thickness is 2mm same thickness for outer cylinder it is 17 mm bigger in diameter than the inner cylinder it holds the inner cylinder allows it to rotate it also has two tabs for left and right motion! thats all for today up next is building up the base to attatch this to

Working on the gimbal small steps understood what a gimbal is through cardboard

Today I made the sensor actually work and pick up data, I did it by using <wire.h> in arduino ide next step is to confirm everything works then add a gimbal mechanism!

I added the Integral and derivative force to the simulator and even added slight wind turbulance to test it against real conditions and its working perfectly, I just bought a IMU sensor MPU6050, Goal of the IMU stage: turn the sensor’s raw, messy readings into one clean tilt angle we can actually trust — solid enough to hand to my PID simulator. In the sim I invented the tilt number out of thin air. Now I will to measure it from the real world,. Closing that gap is the whole stage.

I made 4 timelapses covering my research logs, I researched a Lot and started writing code for the main PID controller, but u may be curious what am I even building##I am building vega a model rocket that can go really high and then land back too!,My goal is to learn PID algorithms which I mostly did today My turtle rocket now balances and calculates the errors too, So its really good as of now next is to optimise it by introducing derivatives. learnt a lot today, from making an inverse pendulum simulator to working Proportional algorithm

AraadhOS — DevlogFrom an empty repo to a working web desktop. Vanilla HTML, CSS, and JavaScript — no frameworks, no build step, no shortcuts.—## Initial commitEvery project starts with a small act of faith: a README.md with two lines in it. No code yet, just a freshly initialized git repo and the decision that this is the thing to build — WebOS, a desktop operating system that lives in a browser tab.## Scaffolding — “made three files”The skeleton: index.html, script.js, and style.css. The JS and CSS were empty, the HTML just 8 lines. The unglamorous-but-essential moment of wiring the three files together — followed by the classic merge commit to reconcile local and remote.## First pixels — “the basic landing page”The first commit where AraadhOS started to look like something: 61 lines of CSS and a boot terminal window with the three traffic-light dots. The aesthetic locked in here too — the moody ayu-dark palette, the gold accent (#E6B450), JetBrains Mono. A landing screen that boots you in rather than just dropping you on a desktop.## Bringing it to life — “basic animations”The first JavaScript showed up and the boot screen started doing things: a typing effect spelling out “Welcome to AraadhOS”, a braille spinner cycling like a real loader, and a blinking ASCII mascot — a little (◕‿◕) face that winks and back. The project stopped being a webpage and became an experience.## The big bang — “v1 of Araadh OS”The monster commit. +1,160 lines. script.js went from a few dozen lines to 767. In one push, AraadhOS became a real desktop environment:- A window system — open, close, drag-by-titlebar, focus-to-front, single-instance apps, cascading windows. Generic and shared, so every app gets it free.- An app registry — the architectural heart. Every app is just { name, icon, render() }. Add an entry and it appears in the dock, the menu bar, and the terminal automatically.- A frosted-glass dock — hover-lift, click-bounce, tooltips, open-app indicators.- The apps — Terminal (a working command interpreter), About, Projects, Files, Calculator, Settings.- Persistent settings — accent color, 12/24-hour clock, UI scale, wallpaper, all saved to localStorage.- A PROFILE object — one place to edit your real content; every app renders from it.Not a prototype — a real, usable thing.## The polish sessionA refinement pass to make the whole thing feel finished:- Squashed a bug — a stray keystroke had broken the clock on startup.- Redesigned the menu-bar — frosted glass to match the dock, a gradient brand chip, restyled buttons, and a clock with a live date.- Built a Notes app — an auto-saving notepad with a live word/character counter, backed by localStorage. It slotted straight in thanks to the app registry.- Removed dead weight — a resume.pdf link that 404’d, and the whoami command (a deliberate “no auth rabbit hole right now” call).- Filled the empty desktop — a grid of desktop icons: single-click to select, double-click to open.## Where it standsIn a handful of commits plus a polish session, AraadhOS went from a two-line README to a browser-based desktop OS with a window manager, a dock, desktop icons, seven working apps, persistent theming, and a boot sequence with real character. The architecture is clean enough that adding the next app is a five-minute job.Next: push the polish work, turn on hosting, and let people boot it.

Devlog: Final Engineering Push – Kinematics, Power Architecture, and Assembly Readiness

Date: June 19, 2026
Status: Design Locked / Fabrication Ready

The past several intensive engineering sessions have brought Bionix from a conceptual CAD model to a fully robust, physically viable electro-mechanical prototype. With the core palm and phalange geometry finalized, power distribution mapped out, and the bill of materials optimized, the project is officially ready for the physical fabrication and assembly phase.

⚙️ Mechanical Engineering & Kinematics

The primary mechanical challenge for a tendon-driven bionic hand is managing friction and preventing catastrophic mechanical failure during repeated flexion cycles. We entirely overhauled the internal routing and joint mechanisms to achieve competition-grade reliability.

• Frictionless Revolute Joints: We stripped out the threaded M2 screws from the phalange joints. The joints have been re-engineered to accept smooth 2.2mm solid brass dowel pins, providing a low-friction, high-strength pivot point that ensures fluid finger movement.
• Optimized Flexion Routing (The Active Grip): To eliminate the “bowstring” effect, we redesigned the entry tunnels on the palm. The four main finger tension lines now enter the palm cavity via perfectly aligned tunnels featuring a -10° downward draft angle. This forces the high-tensile braided line flush against the joint grooves, maximizing servo leverage. All sharp 90-degree channel exits were filleted into smooth “trumpet flares” to prevent the threads from shearing under load.
• Antagonistic Elastic Extension (Passive Return): For the fingers to snap back open instantaneously when the servos release, we implemented an external elastic extensor system. We modeled a Unified Wrist Anchor block at the base of the palm. 1mm shock cord anchors at the fingertips, routes seamlessly over the knuckles, and ties off securely at the base of the palm, providing the exact passive tension required for a lifelike, snappy release.
• Removable Maintenance Hatch: The palm features a custom-traced, screw-mounted top plate with a generous cutout for full thumb articulation. This protects the internal servo array while allowing rapid access for tension-line adjustments.

⚡ Electronics & Power Architecture

Running five high-torque micro servos simultaneously while reading delicate analog muscle signals requires isolating the high-current demands of the actuators from the logic board.

• Main Controller: The system is governed by a Raspberry Pi Pico (RP2040), chosen for its robust hardware PWM processing capabilities.
• Power Distribution: A 7.4V 2S Li-ion battery serves as the primary power source. This is routed through a high-efficiency 5V Buck Converter (rated for ≥3 Amps) directly to the servo power rails. The Pico is powered safely without having to sink the massive stall currents generated when the hand grips an object. A 1000uF decoupling capacitor bridges the 5V rail to smooth out transient voltage spikes.
• Biometric Input: We integrated the MyoWare 2.0 muscle sensor, passing raw electromyography (EMG) signals directly to the Pico’s ADC0 (GPIO26) pin.

devlog 5 or 4 i honestly dont remember lol: I am done with the Palm!!!! and I also assembled the fingers to check !!! yes lets goo!! Printing the index finger tomorrow

Devlog 4: I was already done with 2 fingers earlier, I made the 3rd one today and I used Ai to build the pinky and thumb finger as it was just copying the dimensions editing the sketch and fixing the hole problem

What I set out to build:
I wanted to complete the fingers in this session which I did achieve + I am done with palm’s basic layout and structure planning, In next session I will be improving and finalising the palm assembling all of it together for once to check it all fits then work on the forearm

Devlog 3 : Bionix

Ring finger design has been done and assembled in a sepearate folder I did not face any problems while making this design, Only pinky finger and thumb are left then I will start with palm and forearm respectively!

Devlog2 : Unbound bionix,

Update: last devlog I was finished with the index finger dimensions well I just printed index finger to confirm movement and it’s moving as expected, I also Made the middle finger design by editing sketch of index finger but fixing the hole sliding issue,

failures: I encountered multiple failures planning the design of index finger, but I did make it work and I am really happy for that,

what is left?
designing all fingers and thumb according to their dimensions and then designing the palm and forearm respectively and then moving on to the electronics part

session pic:
here is one session pic where I show the assembled middle finger design, for the middle finger the proximal phalange is 5 cm long, the distal phalange is 2 cm and the middle phalange is 3 cm!

Done adding the library feature!! in researca try it out at https://research-copilot-sigma.vercel.app/library