jpl-x / x_multi_agent Goto Github PK

Dear sir ,
How should I proceed if I only want to perform photometric calibration on my infrared dataset? Does the X library provide relevant functions?

Dear sir ,
when I run the code with my own bag. As shown in the figure,It feels like the photometric calibration is not working.why?how to solve?My device is a mobile vehicle with flir_tau2.
best,

How to save trajectory and compare it with ground-truth?

including the camera intrinsic, extrinsic, and the time difference between image and imu?
Thank you very much.

Hello,

In the README description (and other READMEs in the X Navigation Framework GitHub), it talks about fusing GPS measurements for multi-sensor fusion (see here for an example of a mention to GPS sensor fusion).

In the later part of the README, it gives usage examples for fusing a sun sensor, range sensor, and IMU, but there is no documentation for fusing a GPS sensor. I could not find anywhere in the code that would help point to how to fuse the GPS data.

Could you provide an example on how to fuse GPS data? Or is GPS sensor fusion not supported at this time?

Thank you.

Best,
Curtis

Hi, thanks for open-sourcing the code!
I have some ploblems when running the code.
I have tried the docker environment by the following command: "docker build -t x-image:x-amd64 --build-arg PHOTOMETRIC_CALI=true --build-arg MULTI_UAV=true --build-arg REQUEST_COMM=true ." However, I got the error below:

Sending build context to Docker daemon 76.96MB
Step 1/33 : ARG REPO_NAME="x-image"
Step 2/33 : ARG DESCRIPTION="Docker image for x library"
Step 3/33 : ARG MAINTAINER="Vincenzo Polizzi ([email protected])"
Step 4/33 : ARG ARCH=amd64
Step 5/33 : ARG DISTRO=x
Step 6/33 : ARG BASE_TAG=${DISTRO}-${ARCH}
Step 7/33 : ARG BASE_IMAGE="base-env:${ARCH}"
Step 8/33 : ARG PHOTOMETRIC_CALI=false
Step 9/33 : ARG MULTI_UAV=false
Step 10/33 : ARG REQUEST_COMM=false
Step 11/33 : ARG LAUNCHER=default
Step 12/33 : FROM viciopoli/${BASE_IMAGE} as BASE
pull access denied for viciopoli/base-env, repository does not exist or may require 'docker login': denied: requested access to the resource is denied

I also try to install the dependency required for the code, such as opencv 3.3.1, Boost 1.71.0. I can successfully compile the library and the ROS wrapper. However, when I play the mars_yard_duo_tf.bag, the ros node crash: "[UAV0/x_vio-1] process has died [pid 14110, exit code -11, cmd"

Could you give me suggestions? I would like to use the docker environment. Thank you!

Hi,
I turned the VERBOSE option to ON in the CMakeLists.txt, then the rviz can receive the trajectory topic and tracked feature topic. But, as shown in the following figure, the calibrated thermal images are much brighter than the result in the paper, how can I get the enhanced thermal images as shown in the paper? By the way, what's the black line in the calibrated images?

The calibrated image:

The final estimated trajectory is not bad: