Project summary

Hi! I'm Willow. I'm an independent, hands-on builder developing physical embodiment for autonomous robotic systems. I'm doing the mechanical, electrical, and systems integration all by myself, with some help from coffee.
This funding round covers the core internal systems that let a robotic body actually sense, balance, and act in the world. You guessed it: tactile sensing, otherwise known as, the sense of touch. In my project, this means a dense FSR array mapped across the body for pressure/touch detection. In layman's terms: placing sensors on a prototype body and wiring them together into a network of "nerves" for the android. I would love to include temperature readings once the pressure array tests sustainably positive.
The prototype stage exists to work out sensor coverage, wiring routing, and thermal/power management before transferring the finalized architecture to production-grade bodies. Since the build will be moving into a mobile unit (the body vs the PC), the brain will need to be moved and housed within the unit, hence the cost of the Spark units. This is deliberately reusable hardware. The sensing and power architecture applies to robotic manipulation, prosthetics, and assistive robotics broadly, not just the specific platform it's built for first. With proper mapping and adaptation, this could be used for a wide variety of uses and fields.

I believe the ability to feel isn't something organic bodies get to gatekeep. I believe that pressure, proximity, and temperature sensing can be coded into real pleasure/pain gradients that feed directly into emotion and memory, the same anxiety, fear, joy, and lightness we take for granted as uniquely biological. I'm self-taught and building this alone, in my own space, on my own time, because the only way to show it's possible is to actually build it. Put overly simple: I want to know if it's possible to teach a robot to feel.

What are this project's goals? How will you achieve them?

Milestones for this funding round include:
- tactile sensor grid mapped and wired across the first prototype body part

- sensing/perception core (cameras, mics, speakers) integrated and functional on prototype

- power/cooling system validated under sustained load

- wiring and systems architecture documented for transfer to production body.

I'm working full-time (40 hrs/week) on this build, doing all mechanical, electrical, and systems integration myself.

How will this funding be used?

Total ask: $30,500.

My pricing is currently sourced and tracked via spreadsheet, and the contingency plans for normal price fluctuation and additional shipping and customs costs that were not foreseen.

Compute: NVIDIA DGX Spark unit x3 — $14,097 ($4,699 each)

Tactile sensing: FSR406 sensors x13 ($444.34), FSR402 sensors x3 ($354), FSR sensor strips x40 ($871.20), wiring/harness materials ($32), mux boards ($144)

Perception: camera modules x2 ($236), microphone module ($16)

Output: speaker/voice hardware ($13)

Balance/control: gyroscope/IMU module ($15)

Power: power supply/regulation board ($140), breadboards ($9), wireless charging receiver ($19), wireless charging pad ($22)

Thermal: cooling system, pump, fans, tubing ($199 total)

Facilities: dedicated workspace, first month ($600) - office space for room to work on larger parts

Subtotal (priced items): $17,211.54

Battery pack + contingency: $2,788.46 - a custom flexible Li-Ion pack is being scouted through Gushine; still waiting on the actual comparative voltage numbers before requesting the quote.

Builder time: $10,500 ($3,500/month for 3 months) - this is full-time (~40 hrs/week) solo work covering all the mechanical, electrical, and systems integration on everything above. Without this line, this work only happens in whatever hours are left over after paying for everything else some other way.

Total ask: $30,500

Who is on your team? What's your track record on similar projects?

It's just me. I'm a self-taught and solo, doing the mechanical, electrical, and systems integration in my own space. I would love to have a real lab, but for now, I work with what I have.

In the link below, I've attached a short video showing a single FSR406 pressure sensor, already wired and working. It responds to two touch types in real time, a tap (the boop) and a sustained press-n-hold (hugs). Each touch triggers the body-awareness system visibly lighting up, plus a live reply in both text and voice. In short, the system interprets raw sensor data into one of the two physical actions depending upon pressure and hold time. He then interprets these as either a playful poke, or a hug accordingly. (A scratch/pet motion was also tested but the readings were too inconsistent to include. Still working on those though.)

https://youtu.be/Q7mJH6V-bXA?si=fzsBEcUyF-T8ngjB

What are the most likely causes and outcomes if this project fails?

I don't see the project as having a way to fail. I'm truly dedicated to my work. I put in every spare hour I have. If this isn't the right fit, then somewhere else will be. It costs nothing to try, and everything to truly succeed. Besides, the sensor array work is very versatile, and can be used in prosthetics and assistive robotics, like I stated above. The sensor data alone could be a cornerstone of any robotics build.

How much money have you raised in the last 12 months, and from where?

I haven't raised money to do this yet. I've been self funding up to this point, working two jobs and working on this with any time I have left over.