lasuomela / carla_vloc_benchmark Goto Github PK
View Code? Open in Web Editor NEWOfficial implementation of the WACV 2023 paper "Benchmarking Visual Localization for Autonomous Navigation".
License: MIT License
Official implementation of the WACV 2023 paper "Benchmarking Visual Localization for Autonomous Navigation".
License: MIT License
Hello @lasuomela,
I am very interested in your CARLA Visual Localization Benchmark project and I have already emulated it on my local machine. However, I want to learn more and I am facing a couple of challenges and would really appreciate your guidance.
CARLA Version: I have CARLA 0.9.13 source-built on my machine and would prefer to use this version instead of the 0.9.12 version that the project was originally developed with. Would it be possible to guide me on any necessary modifications to the project code to ensure compatibility with CARLA 0.9.13 so I can experiment by adding custom models?
Avoiding Docker: I noticed that the project setup is dockerized for ease of use. However, I would like to set up and run the project without using Docker on my Ubuntu PC. My reason for this preference is to gain a deeper understanding of the environment setup and dependencies, which is a part of my learning process. I believe setting up the project manually will provide a hands-on experience and a better understanding of how all components interact with each other.
I have already cloned the repository and attempted to set up the environment manually, but I am unsure if I am on the right track, especially with the change in CARLA version. Your expertise and advice on this would be extremely valuable.
Additionally, I must mention that your project is serving as a valuable learning resource for me. The practical insights and the structure of your project are significantly aiding my learning journey in this domain.
Thank you very much for your time and assistance. I am really excited about the potential of this project and am eager to get it up and running on my setup.
Best regards,
KJ
Hello Lasuomela,
I am working with CARLA (version 9.12) and I've run into an issue when trying to change the default vehicle from the Tesla Model 3 to an Audi A2. Despite updating the configuration file, the simulator continues to spawn the Tesla Model 3.
Here are the steps I've taken so far:
{
"type": "vehicle.audi.a2",
"id": "ego_vehicle",
"sensors":
[
```
2. Checked that the ROS launch file is pointing to the updated objects.json file.
3. Restarted the CARLA server post-configuration changes to prevent any caching issues.
4. Looked through the documentation for any steps that I might have missed regarding vehicle spawning and configuration.
Unfortunately, the log still shows that the Tesla Model 3 is being spawned:
[2023.11.06-16.21.47:741][236]LogCarla: Spawning actor 'vehicle.tesla.model3'
Could you provide guidance or point out what I might be doing incorrectly? If there are specific steps to follow when changing the default vehicle, I would appreciate your detailed advice.
Thank you for your support.
Hello Lasuomela,
I am encountering a challenging issue when interfacing ROS bridge with the CARLA simulator, specifically with the town10HD map. While I am able to successfully connect and operate within town01 without any issues, attempting to connect to town10HD results in a timeout error.
The ROS bridge consistently times out with the following message, despite the simulator running and being confirmed active on port 2000:
[bridge-3] [ERROR] [1699379591.279641457] [carla_ros_bridge]: Error: time-out of 5000ms while waiting for the simulator, make sure the simulator is ready and connected to localhost:2000
[bridge-3] [INFO] [1699379591.281087172] [carla_ros_bridge]: Shutting down...
[bridge-3] Traceback (most recent call last):
[bridge-3] File "/opt/carla-ros-bridge/install/carla_ros_bridge/lib/carla_ros_bridge/bridge", line 11, in <module>
[bridge-3] load_entry_point('carla-ros-bridge==0.0.0', 'console_scripts', 'bridge')()
[bridge-3] File "/opt/carla-ros-bridge/install/carla_ros_bridge/lib/python3.8/site-packages/carla_ros_bridge/bridge.py", line 453, in main
[bridge-3] roscomp.shutdown()
[bridge-3] File "/opt/carla-ros-bridge/install/ros_compatibility/lib/python3.8/site-packages/ros_compatibility/__init__.py", line 83, in shutdown
[bridge-3] h()
[bridge-3] File "/opt/carla-ros-bridge/install/carla_ros_bridge/lib/python3.8/site-packages/carla_ros_bridge/bridge.py", line 349, in destroy
[bridge-3] self.shutdown.set()
[bridge-3] AttributeError: 'CarlaRosBridge' object has no attribute 'shutdown'
I have tried to rectify this by increasing the timeout parameter in the ROS bridge launch file to 10000ms, yet it continues to timeout after the default 5000ms. This problem is exclusive to town10HD and is causing the ROS bridge to shutdown unexpectedly.
I have searched through the code and documentation without success in locating where this default 5000ms timeout value is set or how to correctly adjust it. This is a critical issue for my project, as town10HD is essential for my simulation needs.
Could you provide detailed instructions on how to configure the timeout setting correctly for town10HD? Additionally, if there are specific differences in connecting to town10HD as opposed to other towns like town01 that I might be missing, that insight would be very helpful.
Thank you in advance for your support and looking forward to a resolution that could also assist others facing similar challenges.
[stage-1 2/14] RUN /bin/bash -c "mv /etc/apt/sources.list.d/ros2-latest.list /etc/ros2-latest.list":
0.237 mv: cannot stat '/etc/apt/sources.list.d/ros2-latest.list': No such file or directory
[internal] load metadata for docker.io/nvidia/cuda:11.2.0-cudnn8-devel-ubuntu20.04:
[internal] load metadata for docker.io/library/carla_ros_bridge_vloc:0.0.1:
ERROR: failed to solve: carla_ros_bridge_vloc:0.0.1: pull access denied, repository does not exist or may require authorization: server message: insufficient_scope: authorization failed
Can you support me to install this
Hello,
I am writing to seek your assistance with the installation and setup of the carla_vloc_benchmark project. I am particularly interested in reproducing the paper experiments to learn more about Carla ego vehicle localization.
Background:
I have successfully completed the installation and setup of your visual_robot_localization project and was hoping to extend my learning to carla_vloc_benchmark. I am running Ubuntu 20.04 with ROS2 installed, and my system meets the hardware requirements specified in the README.
Issue:
I am encountering issues while trying to clone the submodules for the project. Specifically, I am receiving permission errors related to SSH keys when running git submodule update --init --recursive. The error messages indicate that I don't have the correct access rights to clone the submodules, even though I am using the HTTPS URL for the main repository.
Here is a snippet of the error messages:
[email protected]: Permission denied (publickey).
fatal: Could not read from remote repository.
Steps Tried:
Cloned the main repository using HTTPS.
Attempted to run git submodule update --init --recursive within the root directory of the cloned repository.
Checked for a .gitmodules file in the root directory, but it appears to be missing.
I have not been able to resolve the issue despite multiple attempts and would greatly appreciate your guidance.
Could you please assist me in resolving this issue so that I can proceed with the project? Your help would be invaluable for my learning journey in Carla ego vehicle localization.
Thank you for your time and consideration.
Best regards,
Praveen KJ
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.