Skip to content
Back to Projects

Multiped Robot — Parkour Navigation

February 1, 2023Completedacademic
robotics matlab kinematics 3d-printing cad

Overview

In this Challenge-Based Learning project (4GA40, Q3 2022-2023), our team of 5 designed, manufactured, and programmed a multiped walking robot capable of navigating a parkour course. The course included a 20-30 degree slope, sharp turns, and gaps that the robot had to traverse as quickly as possible.

Full Report — Multiped Robot
Download

Design Process

Concept Generation

We explored multiple locomotion concepts through an RPC (References-Preferences-Constraints) analysis:

  • Spider-type — 6+ legs, stable but complex
  • Dog-type — 4 legs, faster but less stable
  • Hybrid — combined elements from multiple concepts

The final design was a hybrid between two concepts, optimizing for stability on inclines while minimizing the number of actuators.

CAD and Manufacturing

All custom parts were designed in Siemens NX and manufactured using:

  • SLS 3D printing (Nylon 12) for structural components
  • Laser cutting for flat components

Kinematics and Control

Denavit-Hartenberg Modeling

Each leg was modeled using DH parameters to define the kinematic chain from the body frame to the foot:

Tii−1=[cos⁡θi−sin⁡θicos⁡αisin⁡θisin⁡αiaicos⁡θisin⁡θicos⁡θicos⁡αi−cos⁡θisin⁡αiaisin⁡θi0sin⁡αicos⁡αidi0001]T_i^{i-1} = \begin{bmatrix} \cos\theta_i & -\sin\theta_i\cos\alpha_i & \sin\theta_i\sin\alpha_i & a_i\cos\theta_i \\ \sin\theta_i & \cos\theta_i\cos\alpha_i & -\cos\theta_i\sin\alpha_i & a_i\sin\theta_i \\ 0 & \sin\alpha_i & \cos\alpha_i & d_i \\ 0 & 0 & 0 & 1 \end{bmatrix}

Inverse Kinematics

Given a desired foot position, the inverse kinematics solver computed the required servo angles. This was implemented in MATLAB and used for:

  • Gait planning — creep gait for stability (only one leg moves at a time)
  • Trajectory generation — smooth foot paths for slope climbing
  • Workspace analysis — ensuring reachable positions for all terrain features

Servo Control

The leg servos were driven by MATLAB-generated trajectories, with the gait sequence coordinated to maintain the center of gravity within the support polygon at all times.

Testing and Results

The robot was tested on a scaled parkour and iterated based on performance:

  • Successfully climbed the 20-degree slope
  • Navigated turns by coordinating leg phases
  • Gaps required careful trajectory tuning to ensure foot clearance

What I Learned

This was my first project combining CAD design, manufacturing, kinematics theory, and embedded control into a single system. The challenge of translating textbook DH parameters into a physical walking robot taught me how theoretical models break down in practice — friction, servo backlash, and weight distribution all matter more than expected.

Technologies Used

MATLAB (kinematics, trajectory planning), Siemens NX (CAD), SLS 3D printing (Nylon 12), laser cutting, servo motors