Getting Started

Welcome to CoreMech. This guide will help you set up your first unit and get it operational in minutes.

Prerequisites

  • • CoreMech Unit (CM-100 or later)
  • • CoreMech OS 2.0+ (pre-installed on new units)
  • • Network connection (WiFi or Ethernet for initial setup)
  • • Admin workstation with coremech-cli installed

Installation

Install the CoreMech CLI on your workstation to manage units remotely.

Terminal
# Install via npm
npm install -g @coremech/cli

# Or via curl
curl -fsSL https://get.coremech.dev | sh

# Verify installation
coremech --version
# coremech-cli v2.4.1

After installation, authenticate with your CoreMech account:

Terminal
coremech auth login
# Opens browser for authentication
# ✓ Authenticated as operator@example.com

Hardware Setup

Setting up your CoreMech unit for first operation.

1. Power On

Press and hold the power button for 2 seconds. The status LED will turn amber during boot, then green when ready. Boot time is approximately 45 seconds.

2. Initial Connection

On first boot, the unit creates a WiFi access point for configuration:

Network: CoreMech-XXXX

Password: Printed on unit label

3. Network Configuration

Connect to the unit and configure your production network:

Terminal
# Discover units on network
coremech discover
# Found: CM-0024 (192.168.1.105)

# Configure WiFi
coremech config wifi --ssid "YourNetwork" --password "..."

# Verify connection
coremech ping CM-0024
# ✓ CM-0024 responding (latency: 3ms)

CLI Reference

Complete reference for the coremech command-line interface.

coremech status

Display status of all connected units.

$ coremech status
┌──────────┬────────┬─────────┬──────────┐
│ Unit     │ Status │ Battery │ Location │
├──────────┼────────┼─────────┼──────────┤
│ CM-0024  │ Active │ 87%     │ Zone A   │
│ CM-0025  │ Active │ 92%     │ Zone B   │
│ CM-0026  │ Idle   │ 100%    │ Dock 1   │
└──────────┴────────┴─────────┴──────────┘

coremech mission start

Start a predefined mission.

$ coremech mission start patrol-zone-a
→ Mission initialized
→ Assigning CM-0024, CM-0025
→ Estimated completion: 45m

coremech telemetry

Stream live telemetry from a unit.

$ coremech telemetry CM-0024 --live
[14:23:01] pos: (12.4, 8.2) | speed: 1.2m/s | heading: 45°
[14:23:02] pos: (12.5, 8.3) | speed: 1.2m/s | heading: 44°
[14:23:03] pos: (12.6, 8.4) | speed: 1.1m/s | heading: 43°

REST API

Integrate CoreMech into your systems via the REST API.

Base URL

https://api.coremech.dev/v1

Authentication

All requests require a Bearer token in the Authorization header:

Authorization: Bearer your_api_token

Endpoints

GET /units

List all units in your fleet

GET /units/:id/telemetry

Get real-time telemetry for a specific unit

POST /missions

Create and start a new mission

POST /units/:id/command

Send a direct command to a unit

Fleet Management

Managing multiple units as a coordinated fleet.

Mesh Networking

Units automatically form a mesh network when in range of each other. This enables:

  • • Direct unit-to-unit communication without central server
  • • Self-healing topology when units join/leave
  • • Shared situational awareness across the fleet
  • • Distributed task coordination

Zone Configuration

Define operational zones to organize your deployment:

zones.yaml
zones:
  - name: warehouse-a
    bounds:
      min: [0, 0]
      max: [100, 50]
    assigned_units: [CM-0024, CM-0025]
    
  - name: loading-dock
    bounds:
      min: [100, 0]
      max: [120, 30]
    assigned_units: [CM-0026]

ML Inference

Run machine learning models directly on the CoreMech Edge Unit.

Supported Frameworks

  • • TensorFlow Lite
  • • ONNX Runtime
  • • PyTorch Mobile (via ONNX export)

Deploying Models

Deploy your trained models to units:

# Deploy model to a unit
coremech model deploy ./my-model.onnx --to CM-0024

# Deploy to entire fleet
coremech model deploy ./my-model.onnx --fleet

# Check deployed models
coremech model list CM-0024
# my-model.onnx (v1.2, deployed 2026-01-15)

Performance

NPU Compute

15 TOPS

Typical Inference

<10ms

Max Model Size

500MB

Concurrent Models

Up to 4