Fast Sense X Robotics AI Platform
Compaсt and scalable on-board computer for mobile robotics with rich Edge AI capabilities
Powerful on-board computer bringing scalable Edge AI capabilities to mobile robotics
It consists of single board COM Express module with Intel CPU, set of edge AI accelerators to inference several neural nets on-board in real time and has numerous hardware interfaces to robotic sensors and actuators. AI accelerators are connected using M.2 PCIe interface and can be scaled depending on the task. It is shipped as ready to use device with software examples of running ROS algorithms with integrated neural nets meaningful for robotic applications running in isolated Docker containers.
Fast Sense X
Computing hardware
Fast Sense X Robotics AI Platform with scalable Edge AI capabilities
Algorithms
3D Navigation with obstacle avoidance
AI-powered pose estimation
Data processing algos
Telepresense
AI-powered pose estimation
Data processing algos
Telepresense
Integration with Robotics Platforms
Drones
Wheel based
Wheel based
Specification
General Features (DNN inference hardware)
- Support up to 3 Coral EdgeTPU devices
- Support up to 2 Myriad X devices
- Intel UHD GPU (gen 11)
- CPU* Intel® Atom™ x6425E 3.00 GHz (Burst) 1.8 GHz Clock Quad Core
- RAM* max. 16GB onboard LPDDR4x up to 4.267 MT/s
- I/O 2x USB 3.1, 6x USB 2.0, 6x UART, I2C, 1wire, Display port
- Onboard storage* UFS 2.0 onboard flash up to 64 GB
- Power Consumption less then 24W TDP (CPU/GPU - 12W, Edge AI - 11W)
- Size 30 x 55 x 84 mm (1.18" x 2.17" x 3.31") incl. heat-spreaders
- Preinstalled OS Lubuntu 18.04.5 LTS
- Ready-to-go docker containers for EdgeAI devices and ROS node
- Easy-to-use inference code examples
Get Datasheet
Just type your contacts to download the datasheet
| Features | Benefits | |
|---|---|---|
| Support up to 6 EdgeAI inference devices at the same time (GPU+2xMyriadX+3xEdgeTPU) | DNN can be inferenced in parallel mode without any performance losses | |
| Each EdgeTPU is capable to run MobileNetV2 at 400fps | Significant performance boost for real applications (MobileNetV2 is most popular backbone) | |
| Latest Intel x86 CPU (Elkhart Lake) with large amount of RAM (16GB) | Capable to run heavy computation algorithms for navigation and sensor data analysis | |
| Low power consumption (less than 24W) | Longer battery mode operation, requires simplified heatsink/heat-spreader design | |
| Rich-set of IO interfaces (8xUSB/6xUART/I2C/1W/DP) | Easy integration to any robot | |
| Compact size (30 x 55 x 84 mm (1.18” x 2.17” x 3.31”) incl. heat-spreaders) | Easy integration to any robot | |
| Wide-input power (7V - 35V) | Can be directly connected with 2S to 8S accu | |
| Industrial temp. range support (*depends on used EdgeAI devices) | Can be used in industrial environment | |
| EdgeAI devices are easy upgradable | Support of upcoming Keem Bay VPU and other EdgeAI devices for even better performance |
ROS middleware
AI powered pose estimation
Multiple depth cameras and LIDAR data fusion for obstacle mapping - (map icon or similar)
3D path planning
On-board object recognition
Web based telepresence control
Integration with Robotics Platforms
We can offer customized solution for your robots
Request customization of your robotic platform
Fast Sense X computer is ideal for controlling a robot that needs to process data from different sensors like video cameras with AI algorithms on the edge.
In this demo we are presenting Fast Sense X AI Edge Computer controlling a robot to follow the selected person. Human detection and semantic segmentation AI algorithms are running on-board in real time on edge AI accelerators
Media Coverage
