EMOS EMOS EMOS
  • Automatika Robotics
/

Overview

  • EMOS – The Embodied Operating System
  • Why EMOS

Getting Started

  • Installation
  • Quick Start
  • EMOS Dashboard
  • Running Recipes
  • Plugins
  • Mapping
  • Recipes & Tutorials
  • EMOS CLI
  • Troubleshooting

Cognition

  • Cognition Recipes Overview
  • Conversational Agent
  • Local Models
  • Prompt Engineering
  • Spatio-Temporal Memory
  • GoTo Navigation
  • Tool Calling
  • Decision Models
  • Semantic Routing
  • Complete Agent

Planning & Manipulation

  • Planning & Manipulation Overview
  • Cortex: The Agentic Harness
  • Memory and Cortex
  • Cortex Driving the Full Stack
  • Cortex Writes Its Own Tools
  • Multimodal Planning
  • VLA Manipulation
  • Event-Driven VLA
  • MoveIt Manipulation

Navigation

  • Navigation Overview
  • Simulation Quick Starts
  • Point Navigation
  • Navigating a Real Robot
  • Multi-Waypoint Missions
  • Path Recording & Replay
  • Automated Motion Testing
  • Vision Tracking with RGB
  • Vision Tracking with Depth

Adaptivity & Resilience

  • Adaptivity & Resilience Overview
  • Multiprocessing & Fault Tolerance
  • Self-Healing with Fallbacks
  • Event-Driven Cognition
  • Visualizing the System Graph
  • Internal-State Events
  • External Reflexes
  • Cross-Component Healing
  • Logic Gates & Composed Events
  • Context-Aware Actions
  • Motion Detection

Core Concepts

  • Architecture
  • Components
  • Topics
  • Events & Actions
  • Routines
  • Status & Fallbacks
  • Launcher
  • Robot Plugins
  • Web UI & API

Intelligence Layer

  • EmbodiedAgents
  • AI Components
  • Cortex
  • Memory
  • Inference Clients
  • Models

Navigation Layer

  • Kompass
  • Robot Configuration
  • Global Planner
  • Controller
  • Drive Manager
  • Mapping & Localization
  • Mission Manager
  • Motion Server

Advanced

  • Configuration
  • Extending EMOS
  • Supported Types
  • Navigation Algorithms
  • Benchmarks

API Reference

  • API Reference
    • agents
      • agents.components
        • agents.components.depth_lift
        • agents.components.imagestovideo
      • agents.clients
      • agents.ros
      • agents.vectordbs
      • agents.models
    • kompass
      • kompass.components
        • kompass.components.defaults
        • kompass.components.mission
      • kompass.ros
      • kompass.config
      • kompass.data_types
      • kompass.ui_elements
      • kompass.callbacks
      • kompass.actions
    • ros_sugar
      • ros_sugar.launch
        • ros_sugar.launch.signals
        • ros_sugar.launch.launcher
        • ros_sugar.launch.system_info
      • ros_sugar.config
        • ros_sugar.config.base_config
        • ros_sugar.config.robot
      • ros_sugar.io
        • ros_sugar.io.supported_types
        • ros_sugar.io.callbacks
        • ros_sugar.io.topic
        • ros_sugar.io.ipc
        • ros_sugar.io.datatypes
        • ros_sugar.io.publisher
      • ros_sugar.robot
        • ros_sugar.robot.transports
        • ros_sugar.robot.command
        • ros_sugar.robot.adapters
        • ros_sugar.robot.cli
        • ros_sugar.robot.feedback
        • ros_sugar.robot.mount
        • ros_sugar.robot.registries
        • ros_sugar.robot.shm
        • ros_sugar.robot.plugin
        • ros_sugar.robot.ros_outputs
        • ros_sugar.robot.bus
        • ros_sugar.robot.mapping
        • ros_sugar.robot.process
        • ros_sugar.robot.types
      • ros_sugar.core
        • ros_sugar.core.event
        • ros_sugar.core.routine
        • ros_sugar.core.fallbacks
        • ros_sugar.core.status
        • ros_sugar.core.monitor
        • ros_sugar.core.base_action
        • ros_sugar.core.component
        • ros_sugar.core.action
      • ros_sugar.tf
      • ros_sugar.actions
      • ros_sugar.base_clients
  1. EMOS /
  2. Navigation Overview
View as Markdown Open in ChatGPT Open in Claude

Navigation Overview#

Set up and use Kompass, the EMOS navigation engine – planners, controllers, and motion servers for autonomous mobility.


Simulation Quick Starts

Launch a full navigation stack in simulation with pre-built configs.

Simulation Quick Starts
Point Navigation

Step-by-step stack setup – robot config, planner, controller, motion server.

Point Navigation
Navigating a Real Robot

The same stack on a real quadruped: localization, LiDAR and velocity commands through the robot’s plugin, with the robot’s own events and actions.

Navigating a Real Robot
Multi-Waypoint Missions

A patrol as one action: waypoints picked on the map, a dwell or a signal at each, paused and resumed from the browser.

Multi-Waypoint Missions
Path Recording & Replay

Save successful paths and re-execute them on demand for repeatable patrols.

Path Recording & Replay
Automated Motion Testing

Run open-loop tests and record response data for system identification.

Automated Motion Testing
Vision Tracking with RGB

Follow a moving target using an RGB camera and a vision model.

Vision Tracking with RGB
Vision Tracking with Depth

Extend RGB tracking with depth sensing for more robust following.

Vision Tracking with Depth
Navigating to What the Robot Sees

Send a VLM’s detections straight to the Planner as a goal: “go to the red chair” with no intermediate pose.

Multimodal Planning
Cortex Driving the Stack

An agent that turns free-form goals into navigation goals, short safety-checked moves and recovery.

Cortex Driving the Full Stack
Previous
MoveIt Manipulation
Next
Simulation Quick Starts

2026, Automatika Robotics

Made with Sphinx and Shibuya theme.