AUTinyDane

AUTinyDane: Small Robotic Dog Development Project

An open-source, affordable, and customizable 4-legged robotic dog. Originally developed as a 5th-semester project at the Department of Mechanical and Production Engineering at Aarhus University, this project covers the full development process from conceptual design to a finalized, functional quadruped. Click the image below to see the robot walking:

System Overview

AUTinyDane is designed to be highly accessible, operating on a unified, single-board architecture that handles all high-level planning and low-level execution in an open-loop configuration.

Hardware Architecture

The robot operates entirely on a single-controller setup:

• Main Controller: Raspberry Pi running the full Python-based control stack.
• Actuators: Csl6336hv servos utilizing a unique custom bevel gear transmission.
• Chassis: Hybrid construction utilizing laser-cut plywood and 3D-printed PLA/PETG components, validated through Finite Element Analysis (FEA) for structural integrity.

Software Architecture

The pure Python software stack manages everything from user inputs to gait generation and hardware interfacing:

• Kinematics: Inverse Kinematics engine calculated entirely in Python (Kinematics.py). Optimal leg dimensions were mapped using an analytical solution.
• Gait & Trajectory Planning: Modular planners for different movement phases operating in an open-loop state machine (GaitPlannerV2.py, StancePlannerV2.py, SwingPlannerV2.py).
• Hardware Interfacing: Direct communication protocols for servo control (HardwareInterface.py).
• Teleoperation: Native PlayStation 4 controller support for real-time directional input and gait switching.

Project Structure (Important Files)

AUTinyDane/
├── src/                    # Core robot operational code
│   ├── Kinematics.py       # Inverse kinematics calculations
│   ├── HardwareInterface.py# Serial/I2C communication to Servos
│   ├── Configuration.py    # Physical dimensions and robot parameters
│   └── PS4Controller/      # DualShock 4 input mapping
├── legacy/                 # Original planning modules
│   ├── GaitPlanner.py      # Locomotion state machine and timing
│   ├── StancePlanner.py    # Stance phase calculations
│   ├── SwingPlanner.py     # Foot swing trajectory generation
│   └── gait_simulator.py   # Visualizer for gaits before physical deployment
├── tools/                  # Utility and calibration scripts
│   ├── calibrate_servos.py # Servo offset and range tuning
│   └── ServoNeutral.py     # Script to hold servos at neutral position
├── tests/                  # Module testing
│   └── Test_kinematics.py  # Validation for IK models
├── run_robot.py            # Main execution script
└── requirements.txt        # Python package dependencies

Installation

1. Clone the repository:

   git clone [https://github.com/jenslajordmunk/autinydane.git](https://github.com/jenslajordmunk/autinydane.git)
   cd autinydane
   

2. Install the required Python dependencies (it is recommended to use a virtual environment):

   pip install -r requirements.txt
   

   Run The robot

   Remember to calibrate for the current assembly of the robot in ServoCalibration.py, then: Power on the Raspberry Pi system, connect the PS4 controller via Bluetooth, and execute the main loop:

   python run_robot.py
   

Link to CAD Model:

View CAD Model in Onshape

License

This project is licensed under the terms of the MIT License. See the LICENSE file for details.

This site is open source. Improve this page.