Skip to content

V2N_SETUP_GUIDE.md

Latest commit

 

History

History
315 lines (241 loc) · 8.84 KB

V2N_SETUP_GUIDE.md

File metadata and controls

315 lines (241 loc) · 8.84 KB

V2N Setup Guide — From Scratch

Complete guide to deploy target_nav on a fresh RZ/V2N board.

Prerequisites

Hardware connected to V2N:

Port Device Purpose
USB → /dev/ttyACM0 Arduino Mega Motor control + encoder feedback
USB → /dev/ttyUSB0 RPLidar A1 2D LiDAR scanner
USB → /dev/video0 Camera RGB video for DRP-AI detection
DSI-1 Touchscreen GUI display (rotated 180°)
wlan0 WiFi module WiFi Access Point

On the V2N board (pre-installed with Renesas BSP):

  • ROS2 Humble at /opt/ros/humble
  • Python 3.12+

On your PC:

  • SSH access to V2N
  • The target_nav project folder

Step 1: Connect to V2N

Connect V2N to your PC via Ethernet. Default IP is 192.168.50.1.

ssh root@192.168.50.1

If this is a brand new board, you may need to set a static IP first.

Step 2: Copy Project to V2N

From your PC (not the V2N):

# Create workspace on V2N
ssh root@192.168.50.1 "mkdir -p /root/ros2_ws/src/target_nav /root/deploy"

# Copy the Python package + ROS2 files
scp -r target_nav/ setup.py setup.cfg package.xml README.md \
       root@192.168.50.1:/root/ros2_ws/src/target_nav/

# Copy launch, config, URDF, resource, scripts, test
scp -r launch/ config/ urdf/ resource/ scripts/ test/ \
       root@192.168.50.1:/root/ros2_ws/src/target_nav/

# Copy DRP-AI binary and model
scp -r deploy/* root@192.168.50.1:/home/root/deploy/

Step 3: Run Full Setup

SSH into the V2N and run the one-shot provisioning script:

ssh root@192.168.50.1
cd /root/ros2_ws/src/target_nav/scripts/setup
chmod +x v2n_setup.sh
./v2n_setup.sh

This script does everything automatically:

  1. Installs Python dependencies (pyserial)
  2. Builds the ROS2 package with colcon build
  3. Deploys DRP-AI binary and libraries to /usr/lib/
  4. Installs helper scripts (gui.sh, launcher.py, remote_desktop.py) to /root/
  5. Sets up WiFi Access Point (SSID: RZV2N_Robot, password: robot1234)
  6. Configures display rotation for DSI touchscreen
  7. Creates and enables 5 systemd services
  8. Checks hardware connections

Options:

./v2n_setup.sh              # Full setup
./v2n_setup.sh --no-build   # Skip colcon build (if already built)
./v2n_setup.sh --wifi-only  # Only set up WiFi AP
./v2n_setup.sh --status     # Check service status

Step 4: Reboot

reboot

On boot, the system automatically starts:

Service What it does
seatd Display manager (must start first)
wifi-ap WiFi AP at 192.168.50.1 (SSID: RZV2N_Robot)
robot Bringup (LiDAR, Arduino, Odometry)
target-nav-launcher Start/Stop GUI button on touchscreen
remote-desktop Web remote at http://192.168.50.1:8080

Step 5: Connect and Use

From the touchscreen

Tap "Start" on the launcher to open the GUI.

From your PC (WiFi)

  1. Connect to WiFi: RZV2N_Robot (password: robot1234)
  2. Open browser: http://192.168.50.1:8080
  3. You'll see the V2N screen remotely
  4. Click Start to launch the GUI

From your PC (SSH)

ssh root@192.168.50.1
source /opt/ros/humble/setup.bash
source /root/ros2_ws/install/setup.bash

# Check ROS2 topics
ros2 topic list

# Check what's running
ros2 node list

# View LiDAR data
ros2 topic echo /scan --once

# View robot pose
ros2 topic echo /odom --once

What's Running After Boot

┌─────────────────────────────────────────────────┐
│ Infrastructure (launched by bringup.launch.py)   │
│   arduino_driver_node  → motor control (serial)  │
│   odometry_node        → encoder → odom TF       │
│   robot_state_publisher→ URDF → TF tree          │
│   rplidar_node         → LiDAR → /scan           │
└─────────────────────────────────────────────────┘

When GUI starts (user clicks "Start"):

┌──────────────────┐  ┌──────────────────┐  ┌──────────────────┐
│ Core 0: GUI      │  │ Core 1: nav_node │  │ Core 2: camera   │
│                  │  │  (nice -5)       │  │                  │
│ GTK3 window      │  │ Navigator        │  │ DRP-AI C++ binary│
│ Map + camera     │  │ Obstacle avoid   │  │ YOLO detection   │
│ Settings panel   │  │ Blind approach   │  │ (Core 3)         │
│                  │  │ Search scan      │  │                  │
│ ←─ NavShmReader  │  │── /cmd_vel ────→ │  │─ DetShmWriter ──→│
│ ←─ LaserShmReader│  │←─ DetShmReader   │  │ /dev/shm frames  │
│ ─→ CmdRingBuffer │  │←─ CmdRingBuffer  │  │                  │
└──────────────────┘  └──────────────────┘  └──────────────────┘

Note: GUI has NO ROS2. All GUI data flows through struct SHM.

Daily Development Workflow

After the initial setup, use sync.sh to push code changes:

# From your PC, in the project directory:
./scripts/sync.sh

This automatically:

  1. Tests SSH connection
  2. Copies all files to V2N
  3. Cleans build directories
  4. Runs colcon build

Then restart the GUI on the V2N (tap Stop → Start, or):

ssh root@192.168.50.1 "systemctl restart robot"

Network Setup

PC (192.168.50.xxx)  ←──WiFi──→  V2N (192.168.50.1)
                                     │
                                     ├── WiFi AP: RZV2N_Robot
                                     │   Password: robot1234
                                     │   DHCP: 192.168.50.100-200
                                     │
                                     ├── SSH: port 22
                                     ├── Remote Desktop: port 8080
                                     └── ROS2 DDS: domain 0

ROS2 topics are accessible from your PC over WiFi:

# On your PC (after connecting to RZV2N_Robot WiFi):
export ROS_DOMAIN_ID=0
ros2 topic list                    # See V2N topics
ros2 topic echo /detections        # See live detections

File Locations on V2N

Path Content
/root/ros2_ws/src/target_nav/ Source code
/root/ros2_ws/install/target_nav/ Built package
/root/deploy/ DRP-AI binary + model
/root/gui.sh Symlink to GUI launcher
/root/start.sh Symlink to bringup script
~/.config/target_nav/calibration.json Saved settings
/var/log/robot_autostart.log Boot log
/dev/shm/v2n_camera Live camera frame (C++ → GUI, mmap)
/dev/shm/v2n_detections C++ detection structs (C++ → Camera process)
/dev/shm/v2n_det Python detection structs (Camera → Nav + GUI)
/dev/shm/v2n_nav Nav state snapshot (Nav → GUI)
/dev/shm/v2n_laser LiDAR points + pose (Nav → GUI)
/dev/shm/v2n_cmd Command ring buffer (GUI → Nav)
/dev/shm/v2n_calibration Calibration params (Camera → C++)

Troubleshooting

Check service status

systemctl status robot
systemctl status wifi-ap
systemctl status target-nav-launcher

View logs

# Robot autostart log
cat /var/log/robot_autostart.log

# Systemd journal (live)
journalctl -u robot -f

# Last 50 lines
journalctl -u robot -n 50

Hardware not detected

# Check USB devices
ls -la /dev/ttyACM0 /dev/ttyUSB0 /dev/video0

# If missing, reconnect USB and check dmesg:
dmesg | tail -20

Rebuild after code changes

cd /root/ros2_ws
rm -rf build/target_nav install/target_nav
source /opt/ros/humble/setup.bash
colcon build --packages-select target_nav --symlink-install

Stop everything

systemctl stop robot target-nav-launcher remote-desktop
pkill -f "ros2 launch"
pkill -f main_gui
pkill -f app_yolo_cam

Reset calibration to defaults

rm ~/.config/target_nav/calibration.json
# Restart GUI — defaults from config.py will be used

WiFi AP not working

systemctl restart wifi-ap
ip addr show wlan0
# Should show 192.168.50.1

Quick Reference

Action Command
SSH in ssh root@192.168.50.1
Start GUI Tap "Start" on screen, or ssh → /root/gui.sh
Stop everything systemctl stop robot
Restart robot systemctl restart robot
View logs journalctl -u robot -f
Sync code from PC ./scripts/sync.sh
Check topics ros2 topic list
Manual bringup ros2 launch target_nav bringup.launch.py
Check services systemctl status robot wifi-ap target-nav-launcher