Charles Van Hook Portfolio

Project overview

Stemming from an interest in robotics and autonomous systems during high school, this project aimed to tackle the challenge of creating a robot capable of independently navigating and mapping a complex environment like a school campus using LiDAR technology.

Timeline

4 Years

Team

Solo Project

Role

Lead Engineer

Components

SLAM Algorithms

LiDaR

Awards

STATUS

Incomplete

Tools

Fusion 360

ROS

‍

Design Process

The robot featured a 3D-printed chassis designed for navigating varied campus terrain. A key feature was its wheel configuration: two powered wheels and two passive wheels in a cross orientation. To address potential mobility issues on ledges or rocks, springs were incorporated into the back wheel's design to help maintain ground contact for at least one powered wheel. The LiDAR sensor was strategically mounted on top for a 360-degree view. The design was constrained by using available spare parts.

3D Render/CAD

Software / Electronics

The system integrated a LiDAR sensor for mapping, motor controllers, wheel encoders (for odometry), and an accelerometer/IMU. An Nvidia Jetson Nano served as the primary processing unit running the Robot Operating System (ROS), likely managing tasks previously prototyped on an Arduino. ROS was used for sensor data processing (LiDAR point clouds), SLAM (Simultaneous Localization and Mapping) algorithm execution, sensor fusion (combining wheel odometry and IMU data for localization), and navigation (path planning, obstacle avoidance).

Electronics

SOFTWARE

Challenges

Working within the constraint of using only available spare parts required creative design solutions. Learning and implementing LiDAR technology and point cloud mapping algorithms from scratch was a significant technical hurdle. A specific mechanical challenge was ensuring stable navigation over uneven terrain (ledges, rocks) with the cross-wheel configuration, addressed by adding springs. Integrating multiple sensors and fusing their data accurately within ROS for reliable localization and mapping posed software complexity. Migrating functionality from a simpler Arduino platform to the more powerful but complex ROS/Jetson Nano environment was another transition challenge.

MEDIA

Other Projects