Thambi robot is a humanoid robot, a research title for my Ph.D on Humanoid Robotics and Expert Systems.
Its the World`s First Tamil Humanoid Robot and 2nd Indic Humanoid Robot.
Singapore`s First multilingual robot with 4 languages to listen and respond – Tamil, English, Chinese and Malay – https://voice.thambi.ai/final
(still adding more languages…)
I’m not concerned about that because that would require true artificial intelligence, ie “The Singularity”. The singularity does not currently exist and may never exist. Machines gaining “consciousness” would be a truly spectacular feat of programming that would require a deep understanding of how the human mind works, which nobody currently has or else we would have The Singularity by now – and we don’t have it. I did attempt to figure out how to make a machine that could learn and reprogram itself, however, my delvings and ponderings into this subject matter kept hitting dead ends and kept leading to profound questions for which I could not surmise answers.
I am working as Dev Ops Engineer in a company named FGE Control Pte Ltd, Singapore
Doing my Ph.D as Part Time
No I prefer to remain solo at least for this first robot build because I think that is what makes this project special and makes it stand out. It is far more impressive and inspiring for one person to pull off a project of this magnitude than for a large team of scientists to pull off the same feat. By remaining solo, I believe more people will take an interest and this will make the project and the growth of the fans of the project a greater success overall.
I anticipate that the building of the robot will be completed by 2020 and that the coding of all of the advanced A.I. and capabilities will be completed by 2022.
I anticipate the build will cost me between $20,000 and $30,000 in raw materials. The rest of the cost is my time and opportunity cost. However, I think those will be VERY well compensated for before this project is through!
The balance is primarily a sensor, programming, and servo coordination challenge and will be a very challenging and FUN thing for me to take on. I simply CANNOT WAIT! But to explain this in depth would take a 50 page essay lol…
The joints will be actuated through a tendon based system powered by servos and a complex pulley system.
The exoskeleton mesh defines the shape of the muscles of the body and will be used as a framework onto which the realistic silicone skin will be built. The hollow area formed by the exoskeleton will be where I will locate all of the electronics, motors, and tendon pulley systems. I create the mesh prior to setting up the bones and electronics in order to have a detailed volume defined beforehand which enables me to decide where to locate the various components so that they will fit appropriately. The mesh will also be designed to deform, rotate, twist, bulge, and extend in a fashion that simulates the movements of muscle beneath the skin. This will make the robot appear to have muscles working inside of it!
Sonar, webcam eyes, pressure sensors (aka strain gauges), gyro-meters, accelerometers.
For large muscle power, it will use custom servos built from DeWalt drill motors.
For medium muscle power, it will use MG996R servo motors which will be customized – can also group these to have multiple servos actuating a single joint via multiple “tendons”.
For small muscle power, SG90 micro servos – for low torque stuff like facial expressions and fingertips.
I am using Java, Keil C, Python, MRL (MyRobotLab) and few more technologies
To build a robot is my dream project. And when i had a confusion in selecting the title, i decided to go for my dream project even though that time there is not much awareness on AI related services
I have been fascinated with robotics and automation since I first started hearing about advanced robotics in my college years. Since then, robotics have continued to grow in our pop culture and in complexity. They have so many incredible robotics videos on Youtube and on movies these days that I got really inspired and decided to finally make my dream of building a robot a reality!
Yes. It will use LiFe batteries. That said, I’m aiming to have most of the robot’s run-time be based on direct feed from a wall electrical outlet. My plan is to have him always plugged in unless he is switching to a new room/electrical outlet in which case he will unplug his main power cable from the wall, walk to the next room/outlet, plug his main power cord into that new outlet and proceed with his business.
Yes, I will be programming on live-stream the entire ai engine myself and the ai will be the most extensive and time consuming part of the project.
True ai or “consciousness” does not currently exist IRL and if it did, it would become “The Singularity”. That’s a scary thing and I do not plan to create that. But who knows maybe I will on accident and the robot will kill me and proceed to take over the world 🙁
I am also using Rasberry Pi . I’m planning to use a full blown proper pc processor RISC V . I’m planning to use a laptop, mini fanless pc, or tablet to be the “brains” running the main software engine in the chest of the robot. This will then interface by USB with a series of Arduinos, Nervo board from inmoov.fr, molex sensor boards and servomotor controller boards. I am not using Pickaxe now, planning to use if needed for some specific modules in future.