Checkpoint Project

Autonomous Warehouse Robot

The Construct Robotics Masterclass

Completed a three-phase automation process for the RB-1 robot, progressing from hardcoded navigation to full autonomy using SLAM and advanced path planning.

Project Overview

This checkpoint project focused on progressively increasing the autonomy of the RB-1 warehouse robot to navigate and handle objects in a simulated warehouse environment. The project evolved through three distinct phases, each building upon the previous one to achieve full autonomous operation.

Starting from basic hardcoded navigation, the robot gradually developed advanced capabilities including simultaneous localization and mapping (SLAM), obstacle detection, and fully autonomous path planning - demonstrating the complete progression from manual control to intelligent autonomy.

My Role

I developed the navigation and mapping software for all three phases of the project, implementing and tuning algorithms for each level of autonomy. This included:

Three-Phase Development Process

1

Phase 1: Hardcoded Navigation

Foundation - Predefined Path Following

In the initial phase, the robot followed a predefined, hardcoded path to complete its warehouse tasks. The robot was programmed with specific coordinates and commands to:

  • Navigate to a specific shelf location using predetermined waypoints
  • Lift and attach the shelf to the robot base
  • Transport the shelf along a fixed route to the drop-off location
  • Release the shelf at the designated position

This phase established the basic motion control and object manipulation capabilities needed for warehouse operations.

2

Phase 2: SLAM-Based Mapping

Environmental Awareness - Mapping & Localization

The second phase introduced Simultaneous Localization and Mapping (SLAM), enabling the robot to understand its environment dynamically:

  • Real-time mapping of the warehouse environment using laser scanner data
  • Continuous localization to track the robot's position within the generated map
  • Obstacle detection capabilities using sensor fusion
  • Basic path planning to navigate around detected obstacles
  • Map storage and reusability for future navigation tasks

This phase introduced environmental perception, allowing the robot to adapt to changes in the warehouse layout.

3

Phase 3: Full Autonomy

Complete Independence - Autonomous Operation

The final phase achieved complete autonomous operation, where the robot could independently plan and execute warehouse tasks:

  • Autonomous navigation using the previously generated map
  • Dynamic path planning to reach any specified goal location
  • Real-time obstacle avoidance and re-planning capabilities
  • Autonomous shelf identification, pickup, and delivery
  • Complete task execution from start to finish without human intervention

This phase demonstrated true robotic autonomy, with the robot making intelligent decisions to accomplish warehouse logistics tasks.

Technologies Used

ROS (Robot Operating System) SLAM (Gmapping) Path Planning Python C++ Navigation Stack Gazebo Simulation RViz Visualization Laser Scanner Sensor Fusion RB-1 Robot Obstacle Avoidance

Simulation Screenshots

Simulation Environment - RB-1 Warehouse Robot

Screenshots showing robot navigation, SLAM mapping, and autonomous shelf handling

Impact & Learning Outcomes

3
Phases of Autonomy
100%
Autonomous Navigation
Real-time
SLAM Mapping

This project demonstrated a comprehensive understanding of robotic autonomy progression. By developing the robot through three distinct phases, I gained hands-on experience with fundamental robotics concepts:

Back to Portfolio