@Ishanth

Built Linux custom image using Yocto tools. Compiled the source code and loaded onto Raspberry -pi 5 device. The screenshots shows the Linux boot-up logs and the Linux terminal login as root.

Most of the firmware has been updated including the schematics. I am still in the process of making the test and build files because there is some error going on with the test kinematics on my view window.

Sadly smth went wrong with wakatime cuz I havent coded in a while so I just assumed it was turned on.. Unfortunately I only got like a fraction of my actual coding time 😭

The governor of Utah issued a drought emergency for the whole state of Utah in May 2026 due to unprecedented low levels of snow pack in addition to an exceptionally warm winter. The state is experiencing severe drought in all of its 29 counties and extreme drought in 22 of these counties.In response, most homeowners do not modify any watering plans or simply water less but keep their schedules since there isn’t really a way to verify whether the ground in one’s lawn really needs water or not. The irrigation systems are set on time schedules rather than soil conditions which leads to wasteful practices with regard to water usage in a state that cannot afford to do this anymore.This sensor is a small and affordable device with one button that answers the single question: does this particular patch of ground need watering now? It has neither subscription nor application nor installation. Just put the probe into the ground and get an accurate reading.

I am making 10 of these to hand out to friends and family in Utah.

Just got the new CAD complete cuz the old one looked ugly.. I have most of the parts already 3D printed. Right now I am working on the code and some other things.

Made some edits to the mecanum wheels.. They were positioned incorrectly and the wheel was also designed wrong

Yess.. I have finally completed the whole design. This time everything was well thought out. Instead of normal wheels I have decided to use mecanum wheels for omnidirectional movement.

Also there will be two cameras instead of one. The first camera is on the chassis as eyes. The eyes give depth perception so the robot can judge how far away an object is without a depth sensor. The second camera will be on the arm head for object detection.

Electronics and the battery sit low at the back of the chassis as a counterweight to balance the arm mounted at the front, keeping the center of gravity stable even when the arm is fully extended.

The arm has 6 degrees of freedom plus a gripper claw. Serial bus servos give position, voltage, and temperature feedback at every joint, which means the arm always knows exactly where it is, no guessing, no drifting. Inverse kinematics handles all the math of translating “go to point (x, y, z) in space” into the correct angles for all 6 joints simultaneously, so I can give the robot a target position in plain language and it figures out how to get there.

Voice commands flow through a full natural language pipeline: a wake word detector listens constantly for “Hey Dum-E,” then Whisper transcribes what I say, then an LLM interprets the intent and converts it into structured robot commands, things like pick up the red block, move left 10 centimeters, go to the charging station, lift your arm, or grab that.

The full pipeline from voice command to physical grab is: wake word, then speech to text, then LLM intent parsing, then object detection, then coordinate mapping, then inverse kinematics, then servo motion. All of this runs on the Raspberry Pi 5 with an Arduino handling the real-time motor and servo PWM timing underneath, because the Pi’s Linux scheduler isn’t real-time and would cause jittery motion if it controlled servos directly.

Just made some edits.. the UI is fully upgraded and the music bar is easier to use. The site now also fully functions in most browsers (It was only chrome originally).

FocusMate

FocusMate, an AI-powered study assistant designed to improve learning efficiency through computer vision and natural language processing. Built features that analyze webcam input to track user focus, posture, and study engagement in real time. Implemented AI-driven study tools that generate personalized quizzes and summaries based on learning materials. Designed functionality to support both individual and group study sessions with progress tracking and performance insights. Integrated multiple technologies, including machine learning, computer vision, and web development frameworks, to create a comprehensive educational platform.

Devlog #1: Building Dum-E: A Voice-Controlled Mobile Robot Arm.

Hey everyone! I’m building a desktop robot arm inspired by Dum-E from Iron Man the clumsy little arm that Tony Stark keeps yelling at. Except mine is tiny.

The concept is a 4-wheeled mecanum (omni) rover base with a full 5-joint robotic arm mounted on the back, two cameras on the arm head for eyes, and voice control so I can literally talk to it and it responds with movement and personality: beeps, whirs, droopy sad poses when I scold it, excited wiggling when I praise it.

Still in design phase but parts list is locked and I’m about to submit for funding. More updates soon as I start actually building!