@tom can anyone help me with this

@Alex-Kushleyev, any update on whether I can buy 4 of these RGB AR0144s?

The EIS branch is ready for initial tested. Please see the doc file:

https://gitlab.com/voxl-public/voxl-sdk/services/voxl-camera-server/-/blob/eis/voxl2-eis.md

You will need to build the camera server from this branch (and it's best to be able to do that anyway, if experimenting with eis..). Please note that you will need the latest voxl-cross:V4.0 docker image to build the camera server.

The eis branch is sync'ed to latest dev as of today. Added basic params for specifying EIS behavior, as documented (eis_mode, eis_view, eis_follow_rate). Parameter usage is documented as well as basic voxl-camera-server.conf example is provided. Parameter names may change in the future, but the changes will be documented.

Please use the latest IMX412 driver as instructed. For best results, use input resolution of 4040x3040 and output resolution can be anything. including 3840x2160. 4K60 output is supported (h265 strongly recommended), but you will not be able to view that via the voxl-portal.

This release also supports eis on 3 additional MISP channels (can be different resolutions).

We will be focusing next on improving stabilized ROI behavior for smoother output and experimenting more with rolling shutter compensation.

Please try it out, i would be excited to get initial feedback.

Alex

@george-kollamkulam , yes FLIR Lepton 3.5 just inserts into M0157 board which has the socket for the bare Lepton sensor.

Once you install the sensor into the docket, boot up voxl2 and run voxl-lepton-server. It will autogenerate a default config file /etc/modalai/voxl-lepton-server.conf that will look like this.

{ "spi_bus": 14, "spi_speed": 16000000, "en_i2c": false, "i2c_bus": 0, "shutter_mode": "auto", "flow_shutter_s": 20, "closePeriodInFramesInAir": 1, "openPeriodInFramesInAir": 0, "desiredFfcPeriodMsInAir": 360000, "desiredFfcTempDeltaCentiDegInAir": 600, "en_rotate": false }

You will need to change the SPI bus to 0 and , set en_i2c to true and i2c_bus to 4 because this is the configuration for this Hardware arrangement (M0173 + M0157) per the following documentation : https://docs.modalai.com/voxl2-d0014/#downward-range-finder-and-flir-lepton

Then you can run voxl-lepton-server again in foreground or enable the systemd server using systemctl enable voxl-lepton-server and systemctl start voxl-lepton-server . After that you shoud be able to see the Lepton output using voxl-portal.

Alex

@george-kollamkulam

Hi,

ROS2 Melodic requires Ubuntu 20.04 and the VOXL runs Ubuntu 18.04. You could install ROS2 Foxy following this guide. $ apt-get install voxl-ros2-foxy $ source /opt/ros/foxy/local_setup.bash Likewise you can't directly use ROS2 humble on the VOXL because it requires Ubuntu 22.04. We usually run ROS2 inside a docker container. Check out the prebuilt ROS2 images you can have as the base (https://hub.docker.com/_/ros/).

Let me know if that helps!

Ted

@Alex-Kushleyev Thank you! For the moment we have decided to stick with USB since we can fit that into the current electrical architecture, but I appreciate the offer and may take you up on it at some point in the future if our design changes.

Thanks,
Kerry

@bendraper We generally have to activate the SIM card by inserting it into either an android phone or iphone and activate it with the google fi app before using it with VOXL2

@AidanGallagher With an M0130 USB breakout board

M0130: https://www.modalai.com/products/m0130-3?variant=48186331693360

You can use this cable:
MCBL-00085: https://www.modalai.com/products/mcbl-00085?_pos=2&_sid=0c5d87c5e&_ss=r

and get rid of the need for the doodle breakout board

@Alex-Kushleyev

The issue has been resolved. Thank you for your help.

@Andrew-Jue Hi! Could you share how you managed to get your VOXL2 working with motion capture? Did you also use Starling 2 or just a VOXL2 on another platform? It looks like I ran into the same problems as you while trying to get my Starling 2 (also using VOXL2) to position hold with an Optitrack setup

I tried following your troubleshooting steps and replicating your params exactly, and while it seems like it can be toggled into position hold, it doesn't hold its position at all. It just kinda flies off away as soon as I switch it into position hold. I did a screen recording of the outputs of /mavros/vision_pose/pose and /mavros/local_position/pose during flight and noticed that the values for these topics diverges and converges seemingly randomly

Before flight (Left: /mavros/vision_pose/pose, Right: /mavros/local_position/pose)
24e5cb61-9e40-464e-8941-6c60f804f9b4-Screenshot from 2025-06-11 18-24-58.png

During pos hold (Left: /mavros/vision_pose/pose, Right: /mavros/local_position/pose)
73806f5a-aef4-44f7-9cda-90fa7cd39789-image.png

I reviewed the flight logs and it does look like something is commanding it to move away from its position for some reason.
9df2c832-cb27-4fee-af22-ed7b037e9351-image.png

My theory is that the usb wifi on the Starling2 might be introducing some latency that is causing it to fly erratically, although the visual odometry latency seems fine (I think)

f8abeff3-8923-4ead-a9c6-b6a2134ad612-image.png

Do you have any hints as to what the problem might be?

These are the changes I made to switch the Starling 2 to receive data from the Optitrack system:

In /etc/modalai/voxl-vision-hub.conf, changed en_vio to false { "config_file_version": 1, "en_localhost_mavlink_udp": true, "localhost_udp_port_number": 14551, "en_vio": false, "vio_pipe": "qvio", "secondary_vio_pipe": "ov", "vfc_vio_pipe": "ov", "en_reset_vio_if_initialized_inverted": true, "vio_warmup_s": 3, "send_odom_while_failed": true, "horizon_cal_tolerance": 0.5, "en_hitl": false, "offboard_mode": "off", "follow_tag_id": 0, "figure_eight_move_home": true, "robot_radius": 0.300000011920929, "collision_sampling_dt": 0.1, "max_lookahead_distance": 1, "backtrack_seconds": 60, "backtrack_rc_chan": 8, "backtrack_rc_thresh": 1500, "wps_move_home": true, "wps_stride": 0, "wps_timeout": 0, "wps_damp": 1, "wps_vfc_mission": true, "en_tag_fixed_frame": false, "fixed_frame_filter_len": 5, "en_transform_mavlink_pos_setpoints_from_fixed_frame": false, "vfc_rate": 100, "vfc_rc_chan_min": 980, "vfc_rc_chan_max": 2020, "vfc_thrust_ch": 3, "vfc_roll_ch": 1, "vfc_pitch_ch": 2, "vfc_yaw_ch": 4, "vfc_submode_ch": 6, "vfc_alt_mode_rc_min": 0, "vfc_alt_mode_rc_max": 0, "vfc_flow_mode_rc_min": 0, "vfc_flow_mode_rc_max": 0, "vfc_hybrid_flow_mode_rc_min": 0, "vfc_hybrid_flow_mode_rc_max": 0, "vfc_position_mode_rc_min": 0, "vfc_position_mode_rc_max": 2100, "vfc_traj_mode_rc_min": 0, "vfc_traj_mode_rc_max": 0, "vfc_yaw_deadband": 30, "vfc_vxy_deadband": 50, "vfc_vz_deadband": 150, "vfc_min_thrust": 0, "vfc_max_thrust": 0.800000011920929, "vfc_tilt_max": 0.43599998950958252, "vfc_yaw_rate_max": 3, "vfc_thrust_hover": 0.5, "vfc_vz_max": 1, "vfc_kp_z": 5.2899999618530273, "vfc_kd_z": 5.9800000190734863, "vfc_vxy_max": 3, "vfc_kp_xy": 0.63999998569488525, "vfc_kd_xy": 2.559999942779541, "vfc_kp_z_vio": 5.2899999618530273, "vfc_kd_z_vio": 5.9800000190734863, "vfc_kp_xy_vio": 3.2400000095367432, "vfc_kd_xy_vio": 3.9600000381469727, "vfc_w_filt_xy_vio": 10, "vfc_vel_ff_factor_vio": 0.899999976158142, "vfc_xy_acc_limit_vio": 2.5, "vfc_max_z_delta": 3, "vfc_att_transition_time": 0.5, "vfc_stick_move_threshold": 30, "vfc_flow_transition_time": 1, "vfc_q_min": 10, "vfc_points_min": 7, "vfc_en_submode_announcement": 1, "vfc_disable_fallback": false, "vfc_traj_csv": "/data/voxl-vision-hub/traj.csv", "en_voa": true, "voa_upper_bound_m": -0.15000000596046448, "voa_lower_bound_m": 0.15000000596046448, "voa_voa_memory_s": 1, "voa_max_pc_per_fusion": 100, "voa_pie_max_dist_m": 20, "voa_pie_min_dist_m": 0.25, "voa_pie_under_trim_m": 1, "voa_pie_threshold": 3, "voa_send_rate_hz": 20, "voa_pie_slices": 36, "voa_pie_bin_depth_m": 0.15000000596046448, "voa_inputs": [{ "enabled": true, "type": "point_cloud", "input_pipe": "dfs_point_cloud", "frame": "stereo_l", "max_depth": 8, "min_depth": 0.300000011920929, "cell_size": 0.079999998211860657, "threshold": 4, "x_fov_deg": 68, "y_fov_deg": 56, "conf_cutoff": 0 }, { "enabled": true, "type": "point_cloud", "input_pipe": "stereo_front_pc", "frame": "stereo_front_l", "max_depth": 8, "min_depth": 0.300000011920929, "cell_size": 0.079999998211860657, "threshold": 4, "x_fov_deg": 68, "y_fov_deg": 56, "conf_cutoff": 0 }, { "enabled": true, "type": "point_cloud", "input_pipe": "stereo_rear_pc", "frame": "stereo_rear_l", "max_depth": 8, "min_depth": 0.300000011920929, "cell_size": 0.079999998211860657, "threshold": 4, "x_fov_deg": 68, "y_fov_deg": 56, "conf_cutoff": 0 }, { "enabled": true, "type": "tof", "input_pipe": "tof", "frame": "tof", "max_depth": 6, "min_depth": 0.15000000596046448, "cell_size": 0.079999998211860657, "threshold": 3, "x_fov_deg": 106.5, "y_fov_deg": 85.0999984741211, "conf_cutoff": 125 }, { "enabled": true, "type": "rangefinder", "input_pipe": "rangefinders", "frame": "body", "max_depth": 8, "min_depth": 0.300000011920929, "cell_size": 0.079999998211860657, "threshold": 4, "x_fov_deg": 68, "y_fov_deg": 56, "conf_cutoff": 0 }] } In /etc/modalai/voxl-mavlink-server.conf, changed en_external_ap_timesync to 0 { "primary_static_gcs_ip": "192.168.8.10", "secondary_static_gcs_ip": "192.168.8.11", "onboard_port_to_autopilot": 14556, "onboard_port_from_autopilot": 14557, "gcs_port_to_autopilot": 14558, "gcs_port_from_autopilot": 14559, "en_external_uart_ap": false, "autopilot_uart_bus": 1, "autopilot_uart_baudrate": 921600, "autopilot_mission_delay_start": -1, "autopilot_mission_delay_sound": false, "autopilot_mission_notif_dur": 0.1, "en_external_ap_timesync": 0, "en_external_ap_heartbeat": 1, "udp_mtu": 0, "gcs_timeout_s": 4.5 } In vrpn launch file, remapped vrpn_client_node to mavros/vision_pose/pose with 30Hz update frequency Disabled QVIO and ovins Changed PX4 params: EKF2_EV_CTRL to 11, EKF2_EV_QMIN to 0 voxl2:/$ px4-param show ekf2_ev* Symbols: x = used, + = saved, * = unsaved x EKF2_EVA_NOISE [292,568] : 0.1000 x EKF2_EVP_GATE [293,569] : 5.0000 x EKF2_EVP_NOISE [294,570] : 0.1000 x EKF2_EVV_GATE [295,571] : 3.0000 x EKF2_EVV_NOISE [296,572] : 0.1000 x + EKF2_EV_CTRL [297,573] : 11 x EKF2_EV_DELAY [298,574] : 0.0000 x EKF2_EV_HGT_TO [299,575] : 5000 x EKF2_EV_NOISE_MD [300,576] : 0 x EKF2_EV_POS_TO [301,577] : 1000 x EKF2_EV_POS_X [302,578] : 0.0000 x EKF2_EV_POS_Y [303,579] : 0.0000 x EKF2_EV_POS_Z [304,580] : 0.0000 x EKF2_EV_QMIN [305,581] : 0 829/2162 parameters used. Launched mavros with ModalAI's mavros launch script

@Evan-Hertafeld said in Using MPA within voxl-docker-mavsdk-cpp:

libmodal_pipe

@Evan-Hertafeld @Jetson-Nano

You need to share all those files via a volume mount alongside with the /run/mpa directory to allow docker to probe the data and grab from an MPA pipe.

so in your docker run something like -v /usr/include/modal_start_stop.h:/usr/include/modal_start_stop.h needs to be done - the other option is you just download the actual MPA debian inside your docker image and call it a day - then you will have all the necessary headers, libs, binaries, and you just need to share /run/mpa as a volume to access the data.

Zach

@Jetson-Nano you could just use the CPP mavsdk instance to grab the TOF data from a MPA pipe and then leverage that to do your controls via mavsdk (I am assuming just offboard movement). You can connect over port 14551 just ensure you have localhost comms set to true in voxl-vision-hub.

So this wouldn't be a scenario where the GPS is failing entirely, potentially making it in violation of FAA Remote ID, but would just be a weaker signal?

@yardy In order to update the kernel variant without re-installing the whole SDK, please follow these instructions:

cd <voxl-sdk-release>/system-image # reboot voxl2 into fastboot mode adb reboot bootloader # flash VOXL2 kernel variant 0 fastboot flash boot_a m0054-1-var00.0-kernel.img fastboot flash boot_b m0054-1-var00.0-kernel.img fastboot reboot # OR flash VOXL2 kernel variant 1 fastboot flash boot_a m0054-1-var00.1-kernel.img fastboot flash boot_b m0054-1-var00.1-kernel.img fastboot reboot

You can also test the kernel without overwriting it (the original kernel will be retained after voxl2 reboot:

adb reboot bootloader fastboot boot m0054-1-var00.0-kernel.img # OR fastboot boot m0054-1-var00.1-kernel.img

The latter method is useful when you are not sure if the kernel you are about to try actually works (when you are experimenting with building your own kernel) - you can try the new kernel without the risk of temporarily bricking your board.

Alex