Spider Pig - Autonomous hexapod robot

Video:
https://youtu.be/N5_tMeEpJaA

Bachelor thesis project by a group of seven students (applied physics and electrical engineering) at Linköping University, Sweden.

The robot can navigate a simple maze (with both low and high obstacles) autonomously and be controlled manually with an Xbox controller connected to a laptop. It communicates with the laptop over Bluetooth.

​The system is divided into three subsystems (control, sensors and communication), with one microcontroller each. The subsystems communicate via an SPI bus.

Its walking algorithm is based on inverse kinematics. A PD controller is used to keep it centered in corridors.

Main ​hardware:

  • 1x PhantomX AX Metal Hexapod Mark II Kit (including 18x Dynamixel AX-12 servos, excluding the ArbotiX Robocontroller)
  • ​3x ATmega1284p microcontrollers
  • 7x IR distance sensors 
  • 1x ultrasonic distance sensor
  • 1x MPU-6050 IMU 
  • LCD display
  • FireFly Bluetooth modem
  • Raspberry Pi 3 Model B
  • Raspberry Pi Camera Board 5 MP

This is a companion discussion topic for the original entry at https://community.robotshop.com/robots/show/spider-pig-autonomous-hexapod-robot

Inverse Kinematics

Did you use the IK available or design your own? Can you show a video of it walking?

We’re not using any code

We’re not using any code available for the robot kit, we derived our own formula. See the last ~2 minutes of the linked video for close up footage of it walking (in manual mode).  

IK Algorithm & Terrain Adaptation

To confirm, you used the IR sensors for terrain adaptation, and the walking algorithm adapted accordingly? Really well done. Would certainly like to know more about that (point system versus real-time equations?).

Spider Pig

Inspiration for the name?

https://www.youtube.com/watch?v=714-Ioa4XQw