-
Notifications
You must be signed in to change notification settings - Fork 109
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
roslaunch air_slam vo_euroc.launch遇到了问题 #162
Comments
@DeepestLover3685 你好,应该是环境没配好,TensorRT在转模型时没成功 |
感谢,我的环境是 ubuntu20.04 tensorrt8.61 cuda12.1 我catkin_make AirSLAM时并没有问题 |
可以把这一行改成 setReportableSeverity(Logger::Severity::kINFO); 输出更多些信息试试 |
[11/02/2024-16:08:37] [I] [TRT] [MemUsageChange] Init CUDA: CPU +14, GPU +0, now: CPU 35, GPU 67 (MiB) |
可以 给出完整的log吗? |
roslaunch-llk-27963.log |
这些是ros的log,屏幕种应该还有tensorRT相关的输出,也贴出来吧 |
(airgs) llk@llk:~/catkin_ws/src/air_slam/launch/visual_odometry$ roslaunch air_slam vo_euroc.launch started roslaunch server http://llk:42223/ SUMMARYPARAMETERS
NODES ROS_MASTER_URI=http://localhost:11311 process[visual_odometry-1]: started with pid [27977] |
从输出没能看出来什么问题,你用docker能跑通吗? |
我还没有用过docker,现在就去试一下看看。我还需要更改TensorRT版本到8.5么?谢谢 |
不需要的,AirVO用的是8.5,AirSLAM用8.6。另外,docker里面tensorRT已经配好8.6了 |
你是什么显卡? |
NVIDIA GeForce GTX 1650 Ti |
那可能是显卡的问题,我们之前也有运到过TensorRT8.6在1xxx系列的显卡上不太兼容的问题。你可以先试试docker。我们目前也在开发一个纯点特征的系统,会降低对TensorRT版本的要求,以能够在更多GPU上运行 |
好的,辛苦了,十分感谢解答,加油! |
I encountered the same problem, and found a solution that worked well in Docker. Here’s what helped:
after that I tried again $ roslaunch air_slam vo_euroc.launch |
@xukuanHIT 您好!我在roslaunch air_slam vo_euroc.launch时遇到了错误。
首先是rviz没有显示
终端有这样的报错
Erron lightglue building
Error in SuperPoint building
这是我vo_euroc.launch的配置以及dataroot路径
The text was updated successfully, but these errors were encountered: