Robosapien

I, along with my teammates, Amartya Jha, Annem Saad, Krishna Mittal, and Siddhant Singh, participated in the AI Buildathon by the Masters Union.  The objective of the Buildathon was to create a real AI product solving a real-world problem. And in 2 days, we went from an idea to a working prototype called “Fraudly”, an AI-powered app designed to fight digital fraud among teenagers. Teenagers are among the most vulnerable to digital fraud — yet almost no engaging solutions exist for them. With UPI fraud rising sharply and millions of teens coming online every year, the gap was obvious. What if we could teach fraud awareness by letting users experience scams safely? Thus, “Fraudly” was born. The App simulates real-world scam scenarios (UPI, WhatsApp, SMS), lets users swipe to decide: scam or legit, uses AI to explain red flags instantly, and adapts difficulty based on user behavior. Using tools like Replit and Claude, creating a functional app was possible in no time. The Buildathon f...

Personal Takeaways My personal takeaways from the workshop “How to Build an Autonomous Robot” at IIT Delhi: 1. I Underestimated Myself I initially thought I wouldn’t fit into such a high-level environment. But engineering doesn’t care about labels—it rewards curiosity and effort. 2. Learning Happens Fastest When It’s Hands-On You can watch 10 tutorials on robotics. Or spend 2 days building one—and learn more. 3. Collaboration is Everything Working with experienced students accelerated my learning curve massively. 4. Engineering is Iteration Nothing worked perfectly the first time—and that’s the point. 5. Skills I gained Improved CAD modeling Stronger electronics fundamentals Practical understanding of robot architecture Exposure to ROS and automation systems Real-world problem-solving under constraints Final Thoughts This workshop was intense, messy, and incredibly exciting. It wasn’t about building a perfect robot—it was about understanding the process behind building any robot. And t...

 If I have to break the entire experience of the workshop “How to Build an Autonomous Robot” at IIT Delhi, into a framework for building Robots, it would consist of the following steps: Component Selection, CAD Modeling, Circuit Design, Fabrication, Procurement, Assembly, Manual Mode Coding, Manual Testing, ROS Setup, Autonomous Coding, and Autonomous Testing. 1. Component Selection Every robot begins with constraints—size, weight, task, and environment. This step involves selecting the right motors, sensors, controllers, and structural elements to meet those constraints efficiently. 2. CAD Modeling Before building anything physically, the robot is designed virtually. A complete CAD model helps visualize the structure, validate dimensions, and avoid costly mistakes during fabrication. 3. Circuit Design This is where the robot’s nervous system is planned. You define how power flows, how components connect, and how signals are transmitted between sensors, controllers, and actuators. ...

I attended a workshop “How to Build an Autonomous Robot” at the Central Manufacturing Facility (CMF), IIT Delhi. I walked into the workshop thinking I might be out of place, being the only school kid in a group of engineering students. I walked out enriched and confident. The workshop was led by two very senior professors of the Deptt of Mechanical Engineering, Prof Subir K Saha and Prof Sunil Jha. It wasn’t just a lecture series; it was a full-stack engineering sprint. From design to fabrication, electronics to automation, this was engineering in its purest, hands-on form. What made the experience even better was learning directly from IIT professors and working alongside undergraduate students who were actively preparing for Robocon 2026. Day 1: From Idea to Physical Reality Day 1 was about the physicality of robots, component selection, design, and fabrication. 1. Design Presentations:  The workshop kicked off with design presentations. The group was divided into teams at the ve...

The Inspiration Behind My Robo Hand Assist for Rehab Project! Some incidents have a deep impact on you. In December 2025, someone very close to me met with a horrible accident, suffered multiple injuries in the hands, legs, and hips, had to go through multiple surgeries, a long period in the hospital, followed by a long period of recovery. Among the injuries, one was particularly brutal, not because it was life-threatening, but because it took away control. He suffered a radial nerve injury, a condition that affects the ability to extend the wrist and fingers. In simple terms, the hand loses its ability to open and lift. This leads to something known as “wrist drop”, where the hand is unable to perform even the most basic actions. Holding a bottle. Picking up a pen. Shaking someone’s hand.Movements we never think twice about suddenly become impossible. What followed was a series of complex medical procedures - fracture fixation, nerve repair, and eventually a tendon transfer surgery. F...

 With the hardware done, the next step was the firmware and connecting it to Blynk. The code runs in Arduino IDE and handles three things: Wi-Fi connection, receiving RGB values from the Blynk dashboard, and writing those to the NeoPixel ring. The Blynk app sends values through the cloud to the ESP32, which updates all 16 LEDs. Setting up BlynkIOT: create a project in the Blynk web console, add an ESP32 device, get your Auth Token, and paste it into your code. The phone dashboard is three sliders for Red, Green, and Blue, each linked to a virtual pin.  Testing was plug-and-play. Moved the sliders, the LEDs responded. Ran through a few colors: blue, green, red, and mixes. No issues at all. The lamp also diffuses light well when placed face down on a surface. The acrylic scatters it into a soft glow rather than 16 separate points. It wasn't planned, but it's a useful effect. The working video is up on YouTube: https://youtube.com/shorts/1-iBQc89kcM , and the code is up ...

I wanted to build something at the intersection of IoT and home lighting — something you'd actually keep on your desk. The Mood Lamp ended up being a 10 cm circular lamp with 16 RGB LEDs, controlled from your phone over Wi-Fi. The main component is an Adafruit NeoPixel Ring, 16 x 5050 RGB LEDs, each individually addressable. You push an RGB value to each LED to set the colour. That's what gives you basically infinite color options rather than a fixed set. For the microcontroller, I went with ESP32. The reason: it has Wi-Fi built in. No extra module, no extra wiring. You flash the code from Arduino IDE via micro USB, connect to your network, and the hardware side is essentially done. Three wires connect the NeoPixel ring to the ESP32: data to GPIO4, 5V to VIN, and ground to ground. The body is laser-cut acrylic, circular, about 10 cm across. It holds the ring and the ESP32 in place. I had it cut to fit both parts without needing any adhesive. The phone side runs on BlynkIOT....

With Autopilot and Full-Self Driving (FSD) options, Tesla (Models 3, Y, S, X), cars are one of the most sophisticated and accurate Driverless cars on the road, and continually getting better and better. Rivian (R1S, R1T), the Waymo fleet of all-electric Jaguar I-PACE SUVs with Google’s autonomous driver technology are the others among the notable names. A milestone in the journey of Driverless cars, though, was the 2,850-mile, no-hands road trip called “No Hands Across America”, taken up by two CMU folks, in an almost completely autonomous car in 1995. In July 1995, Dean Pomerleau and Todd Jochem of CMU’s Robotics Institute took an epic, 2,850-mile journey from Pittsburgh to San Diego, in a 1990 Pontiac Trans Sport minivan. Their driver for more than 98 percent of the journey was a computer named the Rapidly Adapting Lateral Position Handler (RALPH), and their minivan was Navlab 5—the latest in a series of autonomous vehicles that had been developed at CMU’s Robotics Institute since 19...

In 2026, Tesla Optimus has become the de facto face of the new age Autonomous Robot revolution. Optimus is a 1.73 m, 57 kg, general-purpose humanoid robot designed for autonomous, repetitive, or dangerous tasks. It utilizes Tesla's EV battery technology, has end-to-end neural network AI capability based on Tesla FSD, and the ability to learn tasks by observing humans. And though it is difficult to fathom, the first humanoid robot belongs to the 1960s, a full 60 years prior to 2026. “Shakey” is the first mobile robot with the ability to perceive and reason about its surroundings. It was physically mobile, and had elementary computer vision, and basic navigation capability.  Shakey (https://www.sri.com/hoi/shakey-the-robot) was created from 1966-72 by the Artificial Intelligence Center at Stanford Research Institute (now SRI International). Shakey could perform tasks that required planning, route-finding, and the rearranging of simple objects. Shakey could perceive its surroundings, ...

Long Post Alert!  IIT Delhi Tryst 2026   https://tryst-iitdelhi.org  I had the opportunity to visit IIT Delhi for the Annual Robotics festival TRYST 2026, and didn't miss it even though it was right in the middle of my final exams. A few days' break in the exam schedule due to Holi helped as well. Though I only had a 3-hour window, I got the vibe: the participants, the energy, the displays, the seminars, and the competitions. I liked the following:  1. The ISRO Exhibition, with models of Chandrayaan-3 (1:15), GSLV Mk III (1:20) and PSLV (1:20). Chandrayaan-3 had achieved India's historic soft landing on the Moon's South Pole on August 23, 2023, making India the first nation to do so and the fourth country overall to land on the Moon. The mission successfully deployed the Vikram lander and Pragyan rover, completing all its scientific objectives, and demonstrating India's advanced space capabilities. The GSLV Mk III, now redesignated as LVM3 (Launch Vehicle Mark-3), i...

Update: completion of my Udemy course 'Biomechanics: The Physics of Human Movement'. The course covered the following:  1. Basic terminology of Biomechanics 2. Physics Concepts: Mechanics  3. Statics of the Human Body 4. The Skeletal System 5. Stability and Resistance of the Human Body  6. The Muscular System 7. Kinematics & Kinetics of Human Movement 8. Motion Analysis

With the Biomechanics course completed, I moved on to a new project. In this and the next few blogs, I will discuss the same. The project is ‘Robotic Hand Assist for Rehabilitation’ for rehabilitation through robotic actuation using computer vision. The idea behind the project is to build a robotic hand-assist, which identifies objects using computer vision and helps a human hand pick and hold the object. This can help patients recovering from nerve damage affecting their hand movement. The robotic device can be strapped on the human hand using velcro, and has the following components - 1. A 3D printed human hand-like structure with velcro 2. A microcontroller 3. A camera for enabling computer vision 4. A servo motor with strings attached for enabling movement 5. Machine Learning component - a CV library Further details on the project in the next few blog posts...

Here is the muscle map of human body. Credit: Udemy Course ' Biomechanics: The Physics Of Human Movement' by 'Emil Cordes' 

Here is the bone map of human body. Credit: Udemy Course ' Biomechanics: The Physics Of Human Movement' by 'Emil Cordes' 

Joints in the human body can be Uniaxial, Biaxial or Triaxial.  Uniaxial joints provide a single degree of rotational freedom. Hinge joints  (elbows, fingers) &  pivot joints (ulna & spoke, cervical vertebra) are uniaxial joints. Biaxial joints provide  two degrees of rotational or translational freedom .  Saddle joints (thumb), Plane joints (ankles, ribs) & condyloid joints (wrist) are biaxial. Triaxial joints provide all  three degrees of rotational freedom. Ball and socket joints (hip, shoulder) are triaxial. Credit: Udemy Course ' Biomechanics: The Physics Of Human Movement' by 'Emil Cordes' 

There are five types of bones in the human body Long tubular bones, example Femur Flat bones, example Sternum Short bones, example Lateral cuneiform, Intermediate cuneiform, Medial cuneiform Irregular bones, example Vertebra Sesamoid bones, example Patella Credit: Udemy Course 'Biomechanics: The Physics Of Human Movement' by Emil Cordes'

In Biomechanics,  we do not treat the human body as one object with the complete weight being applied at the center of gravity. W e often need to approximate the percentage of body weight for different body parts, to calculate the respective forces acting at different points in the body.  Here is a useful reference to the  center of masses of specific body parts, and percentage different parts weigh, in relation to the overall body weight. Diagram Credit: Udemy Course ' Biomechanics: The Physics Of Human Movement' by 'Emil Cordes'  

Let's talk about the statics of a human arm holding a weight. Let's try to calculate how many Newtons of Force is the biceps tendon carrying, and what is the moment in the shoulder joint? We assume that a person of 100 kg is holding a dumbbell weighing 10 kg.  Thus, weight of the upper arm = approx 4% of body weight = 4 kg,  and weight of the fore arm = approx 3% of body weight = 3 kg.  We take the length of the fore arm = 30 cm. Also, for keeping calculations simple, we take g = 10m/s2. Cutting the system free F w1  = 40 N, F w2  = 30 N, F wd  = 100 N α  =  10 ° d 1  = 5 cm, d 2  = 15 cm, d 3  = 30 cm We can calculate  Force on the Bicep, F B  = 172.62 N Shoulder Moment, M A  = 34.5 Nm This is fun! Credit: Udemy Course 'Biomechanics: The Physics Of Human Movement' by Emil Cordes'