huchaohong's profile - activity

Recently, i am reading robot_localization and want to understand how it works. Previously, i have read and understood code and logics in robot_pose_ekf package.

I am confused about how imu data is processed in robot_localization. It seems to me that if config of differential was set to false, then imu can't work in robot_localization. The orientation of imu in 'world' fixed frame is transformed to orientation of base_link in 'world' fixed frame and set as measurement and send to ekf or ukf. But how can this data in world frame be fused with other data in other frame(such as 'odom' frame) ?

Hope i was wrong in understanding robot_localization and someone can fix me.

robot_localization: how to deal with imu data Recently, i am reading robot_localization and want to understand how it wo

Should i calibrate imu offset to base_link? Recently, i am trying to use imu. I have seen some excellent algorithms whic

Thanks for you detailed explanation, i can understand the design more clearly now.

thanks for your sharing, i will dig more into dds to learn its design.

In Why ROS 2.0, there is design goal of ros2 network:

we want ROS to behave as well as is possible when network connectivity degrades due to loss and/or delay, from poor-quality WiFi to ground-to-space communication links.

I am curious about how ros achieves this goal? If the network is poor, what will ros2 do, discard messages?

No, i am not asking about how publishers, subscribers, etc map onto DDS domain concepts. ROS2 says it supports poor-qual

I know ROS2 uses DDS but i am not familiar with DDS. I don't get any clue from ROS on DDS. I think there should be more

I know ROS2 uses DDS but i am not familiar with DDS. I don't get any clue from ROS on DDS. I think there should be more

how does ros2 implement its network design? In Why rOS 2.0, there is design goal of ros2 network: we want ROS to be

I want to know if resolution of grid map can influence precision of localization. Say, i have a map with 5cm resolution, will localization algorithm be limited to 5cm precision? Will this depend on different algorithm such as ACML and other algorithms ?

About AMCL，i noticed the sensor models (BeamModel, LikelihoodFieldModel) implemented in ROS would limit particle's precision to the resolution of grid map. That implies me that the whole ACML algorithm implemented in ROS would never reach a better precision than the resolution of grid map.

I am new to localization algorithm, so i don't know if i am right.

I am trying to keep tracking the udpates in navigation2 repo. But the codes are developed so heavily that i faced dependency problem that nav2 would compile failed against upstream ros2 repos. The upstream ros2 repo is developing heavily as well and it is composed of many sub repos which make me even harder to fix version compatibility.

I don't know if there is an appropriate method to handle this problem.

Thanks. I finally understand how this works

how to maintain nav2 version compatibility with upstream ros2 repos? I am trying to keep tracking the udpates in navigat

Publish topic: is it thread safe? I always have a doubt about thread safe regarding topic publish in multiple threads. I