Spot Payloads and Accessories
Spot's capabilities expand significantly through a curated ecosystem of payloads and accessories that integrate seamlessly with the robot's onboard systems. These modular components range from compute modules that add processing power, to specialized sensors for thermal imaging, lidar, or communication systems. The payload interface is standardized, allowing rapid swapping between different configurations, and all payloads are recognized and integrated into Spot's control software and user interface.
Image Gallery
Specifications
Compatible Robot
Spot
Total Aggregate Payload Limit
Approximately 14 kg distributed on back rails
Mounting Interface
Standardized T-slot rails with defined mounting patterns
Power Connections
Payload power output dedicated from Spot's battery
Data Interfaces
Ethernet, USB, and serial connections for payload communication
Environmental Rating
Designed to match or exceed Spot's IP54 ingress protection
Official Payload Examples
CORE and CORE I/O compute modules with GPU support, Enhanced Autonomy Payload with lidar and extended compute, Thermal imaging packages for electrical and mechanical diagnostics, Communication modules for mesh networking and remote teleoperation, Mounting brackets and cable management systems, Power distribution and expansion modules
Power Budget
Shared from main battery; allocation depends on payload combination
Integration Support
Full API and SDK documentation for custom payload development
Typical Use Cases
High-fidelity plant inspection, SLAM mapping, 3D surveying, thermal diagnostics, long-range remote monitoring, specialized research instrumentation
Key Features
Feature
Standardized payload rail interface enabling quick attachment and removal of different sensor and compute modules
Feature
Compute payloads (CORE, CORE I/O) providing GPU and additional CPU resources for onboard AI and analytics
Feature
Enhanced Autonomy Payload combining lidar and compute for improved navigation and SLAM in complex environments
Feature
Thermal imaging payloads for electrical hot-spot detection, mechanical wear assessment, and facility diagnostics
Feature
Communication payloads enabling long-range connectivity via mesh networking or dedicated radios for remote site deployment
Feature
Multiple inspection packages including panoramic cameras, zoom optics, and specialized spectral imaging for industry-specific needs
Feature
Power and data connectors exposed through I/O modules for custom payload integration and research applications
Feature
Payload health monitoring and configuration visibility through Spot's user interface and APIs
Feature
Partner ecosystem supporting specialized payloads designed for oil and gas, mining, energy, and other verticals
Feature
Documentation and SDK support for organizations developing bespoke payloads tailored to specific applications
Frequently Asked Questions
The total payload capacity is approximately 14 kg distributed across the back rails. Users can mount multiple payloads simultaneously as long as the aggregate weight and power draw remain within this limit.
Compute payloads add onboard GPU and CPU resources, allowing users to run heavy AI inference, 3D reconstruction, or data processing directly on the robot rather than streaming all data off-board, which is critical for low-bandwidth or disconnected deployments.
Enhanced autonomy payloads integrate lidar sensors with additional compute, enabling better simultaneous localization and mapping (SLAM), more robust navigation in GPS-denied environments, and improved autonomous path planning in large or complex facilities.
Yes. Boston Dynamics publishes mechanical and electrical interface specifications and provides SDK support, allowing customers and research partners to design and integrate proprietary payloads tailored to their unique applications.
Official and many partner payloads expose configuration and status through Spot's UI and APIs. Custom payloads can expose telemetry through the same interfaces if developed according to the published integration guidelines.
Heavier payloads and higher compute loads increase power consumption. The typical 90-minute runtime is specified for a lightly loaded configuration. Users should expect reduced runtime proportional to added weight and computational load.