it's been a while (but not without updates!)

I did not have much free time to work on this over the last couple of months,

but every once in a while when I did remember (hard not to, I keep the robot hand right in front of the computer), I tried to do something, mostly in the hardware procurement side of things.

some of the things I ended up acquiring:

- Raspberry Pi Zero (also the Raspberry Pi Zero W, which has wifi+bluetooth internally, but I just received it today thanks to Mr. Arjun Rao), for more versatile control (also to make possible to connect to the Leap Motion)

- AA battery power holders, to use AA batteries to power the servos

- A PCA9685 chip, for multiple servo control

- various electronic circuitry (probably won't be necessary for this project anymore)

in addition, I was made aware of a nice project called Cylon.js and its sister project Gobot, which allows users to program microcontrollers using JavaScript and Golang, respectively.

I really wanted to try Gobot and take a dive into the Golang world, but it currently does not have support for the PCA9685, which Cylon.js seems to provide.

with that said, and with a few lines of code, here is the result:

the control of the servos are done through the PCA9685 connected to a Raspberry Pi Zero (no more Arduino), and the servos are powered through the 4-AA batteries.

the next step is to read values from the Leap Motion, which Cylon.js claims to provide.

when that works, we are starting to reach a good place.

Comments

finally got around to it (nrf24l0+ and servo)

On a previous post , I used the nrf24l01+ wireless chip to communicate between the Raspberry Pi and an Arduino, but only got lights to turn on. I remember being confused as to why servos would not work, and somewhat left it there. I started messing around with it again, and I am concluding that it might have been just a power issue. Here is the servo moving properly: The Arduino is on the ground due to the short length of the wires powering them. Just as a recap, what is happening is: - a C++ program using the RF24 library is compiled in the Raspberry Pi (connected to an nrf24l01+ chip) to broadcast a message. When executed, it will broadcast the message. - the Arduino (connected to another nrf24l01+ chip) programmed to receive messages receives the message, and upon receipt sends a signal to an Arduino that is wired to the servo to move the servo. Two separate Arduinos are used, as it seems that the servo library and the RF24 library do not seem to run properly toget...

alcohol sensor (and some patience)

Soldered the alcohol sensor into something that is connectable: I tried to connect this to the Arduino, as I had the appropriate circuitry, but I did not get any legitimate output from it. 5V going in, 5V coming out with no variations. Nothing seems to be awry in wiring, as the circuit seems to be grounded properly (and the 5V current is flowing). There are a couple of potential factors as to why I'm not seeing any results: - I'm using a 10k ohm resistor, while some guides (and the datasheet for the sensor) asks for 100-200k. However, there seems to be a good amount of people using 10k and getting at least some kind of result. A batch of 100k ohm resistors I ordered is on its way, so I guess I can try with them when they come. - This site claims that these sensors take 24-48 hours for its signals to be stable. It also tells me that I should not be powering the sensor directly from the Arduino, which I have been doing, out of concern that the power draw of ...

how the fingers work

the fingers are moved using wires and tension. each finger is looped through with a wire/string/line in a top-bottom fashion. when the top wire is loosened/bottom wire is tightened, the finger curls in its joints, resulting in a "bent finger". you don't have precise control of each joint movement, but it is simple enough to be controlled by a single actuator (in this case a servo). here is me just using my own hands as an actuator: here it is hooked up to as servo and controlled by a potentiometer: the movement is not smooth because 1) it's controlled by a potentiometer that I am controlling, as seen in the background, and I was unable to move it all the way with one smooth finer motion. 2) the tension in the line was not entirely tight. I need to learn how to tie these well. now I have a get a bunch of servos. much credit and thanks to Inmoov once again.

the quest for robots

Search This Blog