Intro
I started like many, hearing about then getting familiar with the Raspberry Pi, playing with digital and analog circuits controlled via its GPIO pins.
1st Mini Project
From basic circuits, I moved on to some more serious projects, such as object-detection / motion-tracking using acoustic sonar (HCSR04s). (INSERT YOUTUBE LINK HERE).
2nd Mini Project
I bought and built the robot-car from Sunfounder. It was fun to do but as I got deeper into projects, I realized that the car was just too limited.
3rd Mini Project
I envisioned to some day have an autonomous rover for inside my home, that would learn the layout, accept voice commands, and discern images. Sort of a home-brew mobile personal-assistant.(Actually, my initial motivation was to have it track down the cat and get it to stop whining at night)
I realized I was not prepared to start on a "real" robot so I concentrated my efforts on projects having to do with sensing and interaction.
More Projects
The next project after the acoustic tracking, was very similar - to detect and track an object by using infra-red. (INSERT YOUTUBE LINK HERE)
After that, the next three projects all had to do with voice-recognition , interaction, and learning.
Finally, I did a rather quick project using the Raspberry Pi camera and some simple utilities to perform some image comparsions, and that also via training, allowed the system to "recognize" what it saw. (INSERT LINK HERE)
Finally, though, I was impatient to start work on a robot.
I realized I didn't want a toy. Something robust. Very quickly learned that "robust" costs money.
The least expensive tracked robot from Superdroid Robots is over $600 USD, so I went with an Agent 390 from Acrobotics.
I liked the low, large, stable platform provided by the Agent 390. One can potentially add tall sensors such as cameras, speakers, perhap an arm instead of all that being much lower to the floor.
It's a rather large robot platform for indoors but I was thinking about the future and who knows, maybe take it outdoors, too.
- Control method: autonomously controlled by onboard microcontroller and digital wireless connection from Linux workstation
- CPU: Raspberry Pi 3B
- Operating system: Raspbian
- Power source: 12v SLA
- Programming language: C++
- Sensors / input devices: IR, HCSR04 ultrasonic sensor, Raspberry Pi camera module
- Target environment: Indoors - Carpet / tiles, Outdoors
This is a companion discussion topic for the original entry at https://community.robotshop.com/robots/show/agent-390-tracked-robot