NAVIXA

The Future of
Warehouse Delivery
Is Autonomous

0
Autonomous Stages
0
° LiDAR Coverage
0
% Overnight Ops
0
Subsystems
Process

How Navixa Works

Three steps from order to delivery — fully autonomous

Place Order

Operator enters a shelf ID in the dashboard. Command sent instantly to the robot via MQTT over the local network.

Robot Navigates

Navixa builds a real-time map using LiDAR SLAM, navigates to the shelf, identifies the box with AI vision.

Box Delivered

Suction-based lifting picks the box and delivers it to the drop zone. Order marked complete in real time on the dashboard.

Technology Stack

Built on Real Hardware

Ten integrated technologies across four subsystems

LiDAR

Hardware

360° laser scanning, 12 m range, 8000 samples/sec for real-time SLAM mapping

Astra Pro Camera

Hardware

RGB-D depth camera, 8 MP, used for QR code scanning and 3D box detection

Jetson Orin Nano

Hardware

NVIDIA edge AI platform, 8 GB RAM, 40 TOPS — runs the full autonomy stack

Teensy 4.1

Hardware

High-speed ARM Cortex-M7 microcontroller for motor control & suction actuation firmware

ROS2 Humble

Software

Distributed node architecture for navigation, perception, and actuation pipelines

SLAM Navigation

Algorithm

Simultaneous localisation and mapping — Nav2 stack with A* path planning

YOLOv8

AI / ML

Real-time object detection for identifying medicine boxes and shelf locations

Python + Flask

Web Backend

REST API + real-time dashboard; paho-MQTT bridge between web and robot

MQTT (QoS 2)

Protocol

Exactly-once delivery for emergency-stop commands over the local network

OpenCV 4

Vision

Image processing pipeline: colour filtering, contour detection, QR decoding

Behind the Build

Steel, Code & Vision

Ten weeks of fabrication, wiring, and integration

The Team

Our Creative Team

Four disciplines. One robot.

Nathalie Ahmed
Abdalla Elradi
Mahanna Alhumaimi
Zahra Jassim

Nathalie Ahmed

Navigation & SLAM

Mechatronics Engineering

Hover to fan · Click to cycle