The Social Quadruped

Posted on

After running into problems with the design of my original quadruped robot, I scaled it down to a quadruped with two degrees of freedom, which uses 8 servos instead of 12.


The Social Quadruped is completely wireless and is operated by my mobile phone or laptop via a Bluetooth connection. The design is centered around this box:


Another requirement of the class project was that we make use of this box in some way in our project. I chose to completely redesign the box into a shell, so as to save on 3D printing time and reduce its weight. It then served as the head for the Social Quadruped, housing the Arduino, the Arduino sensor shield, the Bluetooth serial module, the ultrasonic sensor, the 9V battery that powers the Arduino, and the 4AA batteries that power the servos. I think it also looks a lot better than the original 😎

The Arduino Sensor Shield V5 greatly simplified my wiring process because it had pins for the servos, ultrasonic sensor, and the Bluetooth serial module. It also enabled me to power the servos separately from the Arduino, and the rest of the sensors.

The legs were composed of two servos, each housed in 3D printed brackets, one that moved from side to side, and the other that moved up and down.

The bracket that makes contact with the ground is large so as to keep the robot balanced when it stands and moves around. The Social Quadruped makes use of a shuffling motion to move, and is capable of being manually controlled to move forward, backward, rotate right, rotate left, and wave. Below is a video demonstrating its movement:

The robot also has an autonomous mode, which causes the robot to walk around without manual control using the ultrasonic sensor to detect objects in its path, wave at the first object it gets close to, and avoid this and any other objects in its path.

First Attempt at a 3 DOF Quadruped

Posted on

For my MAKE class, I decided to create a Quadrupedal robot with three degrees of freedom. The requirements for our project were that our parts had to be designed using a CAD software and the parts had to be 3D printed.

It started off really well, but I started to run into major problems further along the design process. First, my servos did not have enough torque to move the legs in a smooth enough fashion, so I was forced to redesign the legs. Once this was done I was able to get a solid enough understanding of Inverse Kinematics to program the walking motion. This blog was a great resource for learning about Inverse Kinematics:

Unfortunately, once I got it walking I was only able to get it to walk 3 steps before it fell over, so I decided to scale down to a simpler two degree of freedom quadruped for the class. I will definitely pick up this project again over the break. Here is a video of the robot’s walking motion when supported: