Advantech Robotics Internship — Autonomous Mobile Robot

Problem

Advantech needed a working AMR demonstrator capable of autonomous navigation for trade show floors — a platform that could reliably move through unpredictable, crowded environments while showcasing Advantech's edge hardware in a live setting.

TODO: what specific gap this filled — was it a new product line, a hardware showcase, a proof-of-concept for a client?

Approach

• Built out the robot's navigation stack using ROS2, Nav2, and SLAM Toolbox for mapping and path planning
• Iterated through simulation before hardware deployment

TODO: what hardware base — custom chassis, off-the-shelf base platform, sensor suite (LIDAR, depth camera, etc.)

TODO: what Advantech edge compute hardware was integrated — likely central to why it mattered to them

TODO: confirm whether Isaac Sim was used here specifically, or came later with the Aveline project

TODO: the most valuable part of this case study — a specific technical obstacle solved (sensor calibration, localization drift, network/latency on the show floor). Hiring managers want the debugging story, not just the stack list.

Outcome

Demonstrated live at [trade show name(s), city, date] running autonomous navigation in front of attendees.

TODO: any measurable result — uptime across the show, number of demo runs, positive client/visitor feedback, whether it led to further work

To finish this page

A simple stack diagram (sensors → SLAM/localization → Nav2 planner → motor control) plus a sentence or two on what would be done differently now, with current Isaac Sim experience — shows growth rather than just a static achievement.