Robotic Dog

My robot dog is a quadruped designed for agility and balance, powered by LX-16A servos for precise leg movements. It features a carefully planned gait system, enabling it to walk, step, and adapt dynamically to new positions. Designed with a Raspberry Pi as the central controller, the robot is programmed for continuous improvements in motion and stability.

Planning

Before starting this project, I anticipated it would be challenging. To ensure my idea was feasible, I sketched the robot on paper to visualize the design as a whole. Using the specifications from the data sheets of all the components I planned to use, I calculated the optimal weight, limb lengths, and the expected speeds the robot could achieve.

The CAD

After completing the planning phase, I used SolidWorks to create a detailed CAD model of the robot, incorporating all the components into the design.

Assembled Robot

Testing

Once the robot was assembled, I programmed it with basic walking code to test its functionality. This ensured that both the hardware and software were working as planned.

Robot Gait Neural Network

My brother and I are currently working on developing a Robot Gait Neural Network to teach the robot how to walk. The project is still in progress, but our goal is to use this data-driven approach to create smooth and efficient walking motions for the robot. By leveraging machine learning, we aim to optimize the gait for various terrains and conditions.