VnRoboVnRobo
AboutPricingBlogContact
🇻🇳VISign InStart learning
🇻🇳VI
VnRobo logoVnRobo logo

Learn Physical AI and experiment in your browser. Connect your fleet when you have real robots.

Product

  • Features
  • Pricing
  • Knowledge Base
  • Services

Company

  • About Us
  • Blog
  • Contact

Legal

  • Privacy Policy
  • Terms of Service

© 2026 VnRobo. All rights reserved.

Made with♥in Vietnam
VnRoboVnRobo
AboutPricingBlogContact
🇻🇳VISign InStart learning
🇻🇳VI
  1. Home
  2. Blog
  3. Edge AI with NVIDIA Jetson: Deploy AI on Embedded Devices
aiai-perceptionedge-computingnvidia

Edge AI with NVIDIA Jetson: Deploy AI on Embedded Devices

Guide to deploying Edge AI with NVIDIA Jetson and TensorRT — optimize real-time inference for robots and embedded applications.

Nguyen Anh TuanAugust 15, 20253 min readUpdated: Nov 14, 2026
Edge AI with NVIDIA Jetson: Deploy AI on Embedded Devices

What is Edge AI and Why Jetson?

Edge AI means running AI models directly on devices instead of sending to cloud. Advantages: ultra-low latency (<30ms), offline operation, data security, bandwidth savings. In robotics, edge AI is mandatory — robot can't wait 200ms round-trip to cloud to avoid obstacle.

NVIDIA Jetson is SBC line designed for AI inference with integrated CUDA GPUs.

NVIDIA Jetson board and AI embedded devices for robotics
NVIDIA Jetson board and AI embedded devices for robotics

Jetson Product Line

Module GPU Cores RAM AI Performance Price (~)
Jetson Nano 128 CUDA 4GB 472 GFLOPS $149
Jetson Orin Nano 1024 CUDA 8GB 40 TOPS $249
Jetson Orin NX 1024 CUDA 16GB 100 TOPS $599
Jetson AGX Orin 2048 CUDA 64GB 275 TOPS $1999

For mobile robots, Jetson Orin Nano optimal — best performance/price/power (15W).

Environment Setup

JetPack SDK

bash
# Install JetPack (includes CUDA, cuDNN, TensorRT)
sudo apt update &#x26;&#x26; sudo apt install -y nvidia-jetpack

# Verify CUDA
nvcc --version
python3 -c "import torch; print(torch.cuda.is_available())"

Development Container

NVIDIA provides ready-made containers:

bash
# Pull PyTorch container for Jetson
sudo docker pull nvcr.io/nvidia/l4t-pytorch:r36.2.0-pth2.1-py3

# Run with GPU access
sudo docker run -it --runtime nvidia --network host \
  nvcr.io/nvidia/l4t-pytorch:r36.2.0-pth2.1-py3

TensorRT Optimization

TensorRT accelerates models 2-5x vs PyTorch native.

YOLOv8 to TensorRT Conversion

python
from ultralytics import YOLO

model = YOLO("yolov8n.pt")

# Export to TensorRT FP16
model.export(format="engine", device=0, half=True, imgsz=640)

Performance on Jetson Orin Nano

Format FPS (640x640) Latency
PyTorch FP32 12 FPS 83ms
TensorRT FP32 28 FPS 36ms
TensorRT FP16 45 FPS 22ms
TensorRT INT8 62 FPS 16ms

Quantization: FP16 and INT8

FP16 reduces memory half and speeds inference with minimal accuracy loss — default choice for Jetson.

INT8 needs calibration but maximum speed:

python
import tensorrt as trt

config.set_flag(trt.BuilderFlag.INT8)
config.int8_calibrator = EntropyCalibrator(
    calibration_data="/path/to/calib_images/",
    cache_file="calibration.cache"
)

Real-time AI inference on edge computing device
Real-time AI inference on edge computing device

Complete Real-time Pipeline

python
import cv2
from ultralytics import YOLO

model = YOLO("yolov8n.engine")
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()
    if not ret:
        break

    # Inference
    results = model.predict(frame, conf=0.5, verbose=False)

    # Process results
    for box in results[0].boxes:
        cls = int(box.cls[0])
        conf = float(box.conf[0])
        x1, y1, x2, y2 = map(int, box.xyxy[0])

        # Send command to robot based on detection
        if cls == 0:  # person detected
            distance = estimate_distance(y2 - y1)
            if distance &#x3C; 1.0:
                send_stop_command()

Power Management

bash
# View power mode
sudo nvpmodel -q

# MAXN mode (15W, high performance)
sudo nvpmodel -m 0

# Maximize performance
sudo jetson_clocks

Deployment Tips

  1. Always use TensorRT FP16 — baseline, only switch INT8 for extra FPS
  2. Pre-process on GPU: use cv2.cuda instead of CPU OpenCV
  3. Batch inference: group multiple frames if latency allows
  4. Monitor temperature: Jetson throttles at 80°C

Edge AI with Jetson enables deploying powerful AI directly on robots without cloud dependency, meeting strictest real-time requirements. Combined with Python robot programming, build complete AI-to-action pipeline.

Related Articles

  • Computer Vision for Automated Quality Inspection
  • AI Trends in Robotics 2025
  • Deploy IoT Applications with Docker and Docker Compose
NT

Nguyễn Anh Tuấn

Robotics & AI Engineer. Building VnRobo — sharing knowledge about robot learning, VLA models, and automation.

Explore VnRobo

Fleet MonitoringROS 2 IntegrationAMR Solutions
← Previous
Programming Robot Control with Python: From Basics to Advanced
Next →
Micro-ROS: Connecting Microcontrollers to ROS 2 Ecosystem

Related Posts

Deep Dive
Robot AI
ai-perceptionedge-computinggemma
ai

Gemma 4 for Robotics: Open-Source AI Running on the Edge

Deep dive into Google's Gemma 4 — open-source multimodal AI with agentic capabilities, running on Jetson and Raspberry Pi for robotics.

4/12/202610 min read
NT
Robot AI
ai-perceptionnvidiatutorial
ai

Deploy YOLOv8 to Jetson Orin Nano in 30 Minutes

Step-by-step guide deploying YOLOv8 with TensorRT on NVIDIA Jetson Orin Nano — from model export to real-time 60 FPS inference.

2/15/20268 min read
NT
Research
Robot AI
ai-perceptiongemmaedge-ai
ai

Gemma 4 and Its Applications in Robotics

Analyzing Google's Gemma 4 architecture — from on-device AI to practical applications in robot control, perception and agentic workflows.

4/12/202611 min read
NT
VnRobo logoVnRobo logo

Learn Physical AI and experiment in your browser. Connect your fleet when you have real robots.

Product

  • Features
  • Pricing
  • Knowledge Base
  • Services

Company

  • About Us
  • Blog
  • Contact

Legal

  • Privacy Policy
  • Terms of Service

© 2026 VnRobo. All rights reserved.

Made with♥in Vietnam