MyRoboPath
Intermediate
6 - 8 Months
$105,000 - $155,000 / year

Autonomous Mobile Robots (AMR) & AGV Roadmap

Master autonomous navigation: from wheel kinematics and sensor fusion to 2D/3D SLAM and ROS 2 Nav2.

The definitive roadmap for building commercial-grade Autonomous Mobile Robots (AMRs), warehouse Automated Guided Vehicles (AGVs), and delivery robots using LiDAR, wheel encoders, IMU sensor fusion, and the ROS 2 Navigation Stack.

Target Roles:
Autonomous Systems EngineerAMR Navigation EngineerRobotics Perception SpecialistROS 2 Software Developer
Prerequisites:
Proficiency in C++ and PythonBasic understanding of Linux & terminal commandsLinear algebra and matrix arithmetic

Roadmap Curriculum & Milestones

Complete each sequential phase to build production-grade robotics competencies.

3 Major Phases
01

Phase 1: Mobile Robot Kinematics & Odometry

Formulate differential drive, Ackermann steering, and mecanum omnidirectional kinematic equations in code.

Step 1

Differential & Omnidirectional Kinematic Models

Duration: 3 Weeks

Derive forward and inverse kinematics converting linear (v) and angular (w) velocities to individual wheel RPMs.

Core Competencies:
  • Forward/Inverse Wheel Kinematics
  • Dead Reckoning Odometry
  • Runge-Kutta 2nd Order Integration
  • Covariance Estimation
Hands-On Projects:
  • Dead-reckoning odometry publisher node in C++
Tools:ROS 2 Humble / IronEigen3 C++ Linear AlgebraGazebo Simulator
Step 2

Sensor Fusion with Extended Kalman Filter (EKF)

Duration: 3 Weeks

Fuse wheel encoder ticks with 9-axis IMU (gyroscope and accelerometer) using the robot_localization ROS 2 package.

Core Competencies:
  • Extended Kalman Filter (EKF)
  • IMU Drift Compensation
  • robot_localization Package Configuration
  • TF2 Coordinate Tree (odom -> base_link)
Hands-On Projects:
  • Robust filtered odometry publisher fusing BNO055 IMU + Encoders
Tools:robot_localizationBNO055 9-DOF IMURViz 2
02

Phase 2: LiDAR Mapping & 2D/3D SLAM

Generate high-resolution 2D occupancy grids and 3D point clouds in unknown environments in real time.

Step 1

2D LiDAR SLAM with Cartographer & SLAM Toolbox

Duration: 4 Weeks

Implement scan matching, loop closure detection, submap generation, and occupancy grid serialization.

Core Competencies:
  • LaserScan Message Protocol
  • Scan Matching (Correlative & Ceres)
  • Loop Closure Optimization
  • Lifelong Mapping
Hands-On Projects:
  • Mapping a multi-room environment with SLAM Toolbox and RPLiDAR
Tools:SLAM ToolboxGoogle CartographerRPLiDAR A1/C1RViz2
Step 2

Adaptive Monte Carlo Localization (AMCL)

Duration: 3 Weeks

Localize a robot on a pre-built static map using particle filter probabilistic state estimation.

Core Competencies:
  • Particle Filtering (MCL)
  • Likelihood Field Sensor Model
  • Kidnapped Robot Recovery
  • Map-to-Odom Transform Broadcasting
Hands-On Projects:
  • Global localization benchmark on noisy map environments
Tools:Nav2 AMCLCostmap 2DROS 2 Humble
03

Phase 3: Autonomous Path Planning & Nav2 Stack

Deploy industrial navigation with global path planners, dynamic obstacle avoidance, and behavior tree orchestrations.

Step 1

Global & Local Path Planners (A*, Dijkstra, TEB & DWB)

Duration: 4 Weeks

Configure Nav2 layered costmaps (inflation, obstacle, voxel) and tune local trajectory controllers for collision-free motion.

Core Competencies:
  • A* & Dijkstra Pathfinding
  • Timed Elastic Band (TEB) Controller
  • Dynamic Window Approach (DWB)
  • Costmap Layer Plugins
Hands-On Projects:
  • Full autonomous navigation pipeline with waypoint following
Tools:Nav2 StackBehaviorTree.CPPROS 2 Actions

Ready to begin Phase 1?

Dive into our free hands-on tutorials and build your first physical prototype.

Explore Tutorials