|
| 1 | +#!/usr/bin/env python3 |
| 2 | + |
| 3 | +""" |
| 4 | +Send visual odometry data to PX4 via MAVSDK's Mocap plugin. |
| 5 | +
|
| 6 | +This example demonstrates how to send ODOMETRY messages to PX4 for |
| 7 | +vision-based navigation (VIO, visual odometry, motion capture, etc.). |
| 8 | +
|
| 9 | +It connects to a PX4 simulator, configures EKF2 for external vision fusion, |
| 10 | +sends simulated odometry at 30 Hz, and monitors health to verify fusion. |
| 11 | +
|
| 12 | +Prerequisites: |
| 13 | + - PX4 simulator running (e.g. make px4_sitl gz_x500) |
| 14 | + - pip install mavsdk |
| 15 | +""" |
| 16 | + |
| 17 | +import asyncio |
| 18 | +import math |
| 19 | +import time |
| 20 | + |
| 21 | +from mavsdk import System |
| 22 | +from mavsdk.mocap import ( |
| 23 | + Odometry, |
| 24 | + PositionBody, |
| 25 | + Quaternion, |
| 26 | + SpeedBody, |
| 27 | + AngularVelocityBody, |
| 28 | + Covariance, |
| 29 | +) |
| 30 | + |
| 31 | + |
| 32 | +async def run(): |
| 33 | + drone = System() |
| 34 | + await drone.connect(system_address="udp://:14540") |
| 35 | + |
| 36 | + print("Waiting for drone to connect...") |
| 37 | + async for state in drone.core.connection_state(): |
| 38 | + if state.is_connected: |
| 39 | + print("-- Connected to drone!") |
| 40 | + break |
| 41 | + |
| 42 | + # Configure EKF2 for vision fusion |
| 43 | + print("Configuring EKF2 for vision fusion...") |
| 44 | + await drone.param.set_param_int("EKF2_EV_CTRL", 15) |
| 45 | + await drone.param.set_param_int("EKF2_HGT_REF", 3) |
| 46 | + print("-- EKF2 configured (EV_CTRL=15, HGT_REF=3)") |
| 47 | + |
| 48 | + # Start sending odometry and monitoring health in parallel |
| 49 | + await asyncio.gather( |
| 50 | + send_odometry(drone), |
| 51 | + monitor_health(drone), |
| 52 | + ) |
| 53 | + |
| 54 | + |
| 55 | +async def monitor_health(drone): |
| 56 | + """Print health status until vision fusion is active.""" |
| 57 | + async for health in drone.telemetry.health(): |
| 58 | + status = "OK" if health.is_local_position_ok else "WAITING" |
| 59 | + print(f" Local position: {status}") |
| 60 | + if health.is_local_position_ok: |
| 61 | + print(" --> EKF2 is fusing vision data!") |
| 62 | + break |
| 63 | + |
| 64 | + |
| 65 | +async def send_odometry(drone): |
| 66 | + """Send simulated vision odometry at 30 Hz for 20 seconds.""" |
| 67 | + nan_cov = Covariance([float("nan")]) |
| 68 | + start_time = time.monotonic() |
| 69 | + |
| 70 | + print("Sending vision odometry at 30 Hz...") |
| 71 | + for i in range(600): |
| 72 | + elapsed = time.monotonic() - start_time |
| 73 | + |
| 74 | + # Simulated position: gentle circle at 0.5 m radius |
| 75 | + x = 0.5 * math.cos(elapsed * 0.5) |
| 76 | + y = 0.5 * math.sin(elapsed * 0.5) |
| 77 | + z = 0.0 |
| 78 | + |
| 79 | + # Simulated velocity |
| 80 | + vx = -0.25 * math.sin(elapsed * 0.5) |
| 81 | + vy = 0.25 * math.cos(elapsed * 0.5) |
| 82 | + |
| 83 | + odom = Odometry( |
| 84 | + time_usec=0, |
| 85 | + frame_id=Odometry.MavFrame.MOCAP_NED, |
| 86 | + position_body=PositionBody(x, y, z), |
| 87 | + q=Quaternion(1.0, 0.0, 0.0, 0.0), |
| 88 | + speed_body=SpeedBody(vx, vy, 0.0), |
| 89 | + angular_velocity_body=AngularVelocityBody(0.0, 0.0, 0.0), |
| 90 | + pose_covariance=nan_cov, |
| 91 | + velocity_covariance=nan_cov, |
| 92 | + ) |
| 93 | + await drone.mocap.set_odometry(odom) |
| 94 | + |
| 95 | + if i % 90 == 0: |
| 96 | + print(f" t={elapsed:.1f}s pos=({x:.2f}, {y:.2f}, {z:.2f})") |
| 97 | + |
| 98 | + await asyncio.sleep(1.0 / 30) |
| 99 | + |
| 100 | + print("-- Done sending odometry") |
| 101 | + |
| 102 | + |
| 103 | +if __name__ == "__main__": |
| 104 | + asyncio.run(run()) |
0 commit comments