Sensor Fusion in Mobile Autonomous Robot | ROS | IMU+Wheel Odometry | Kalman Fliter | Jetson Nano

[@nishantsharma4465]

Never disappointed after watching you videos bro . Thank you for sharing these practical information .

[@huytran-sd1kr]

Thanks again. i am just starting my ros learning, this is one of the best sources I found

[@thalanayarmuthukumar5472]

Great 👍✅ video explaining sensor fusion by contrasting the robot behavior with and without sensor fusion. Very inspiring 👏

[@Jp_Robotics]

Very helpful Thank you so much! Please do make more tutorials..

[@ganeshkakumanu8136]

Eagerly waited for a video like this thank you.

[@gowllateja]

Very informative. Thanks a lot.

[@kvnptl4400]

Nice video demo and explanation of fusion odometry

[@ranam]

you are doing a great job bro i am very jealous to copy you its a shame to say that but just for fun bro please continue your work you are delivering a great content bro

[@jasirthachaparamban3359]

Good video bro.. Keep uploading this kind of useful video

[@thalanayarmuthukumar5472]

Great to see the sequence of videos covering different topics. Can you share what you plan as the next video and when? Eagerly awaiting

[@Elnasir]

Easy explanation. Many thanks

[@nhatnet479]

hello friends, could you help me explain this question. In the ekf.yaml file. base_link_frame: base_link => this frame got from odom which calculated from encoder? so we have to run the node odom (calculated by encoder) or the node transform between (base_link and odom) which calculated from encoders ?

[@stephenzuccaro1090]

Great video!

[@vinothbose]

Very good Demo. Thanks

[@raghavs6554]

I have a hardware robot built with IMU and Odometry. When I add the topics and covariance matrix in the ekf yaml, hardware robot is able to move based on the cmd_vel (when we give 2dNavGoal in rviz). But in simulation, the robot is not moving. Any idea on this?

[@chelmartin]

Great Video. Question Why did the blue and red odometry circles not overlap in rviz?

[@AnitaSharma-bk4fc]

Sir I am also using imu 6050 integrating with jetson nano for my project but I am facing many issues 1 st one is i am unable to its address and after that getting error like cmake and all that when I am ignoring i2c detect problem but while launching rviz i am not getting map options

[@fransflorenza9708]

can anyone help me pls.. "(. in your launch file there is "remap from odometry/filtered to odom" but in my experience the robot localization package still affected my original odometry when launched even without remap to odom line. and the strange thing is when im using the remap to odom my robot become moving forward continuously in rviz.

[@mayank8387]

This video is really well done. Informative and concise. Thanks for sharing. Are you a student? Because if you are, you must be hella rich lol. That Jetson and lidar sensor alone must've costed you around 20k am I right?

[@roboticsandroslearning8232]

Yes, you are right. This project costs around 20k. But I am not a student and so I am not hella rich yet.

[@christhaliyath]

Was your IMU in ENU coordinate convention?

[@debanikroy7496]

Great video

[@pavancherukuri2824]

Are you using amcl also?

[@daleermac8573]

Hello Sir, Great work. So i have some question what Communication protocol you use to communicate the robot to your pc

[@fernandoalgarin7017]

Hi Sir , I would like to know how you fused the information given by the LiDAR with the fusion of the IMU and wheel Odometry... I'll appreciate any advice!

[@sanjayshekhar7]

Why Jetson Nano(JN), we can also use Raspberry Pi or something which costs less than JN right ? Any particular reason ?

[@mavencheong]

hi i'm trying to implement the same fusing with diff_drive_controller odometry and IMU, however when the wheel stick in the mat, the position of the odomentry still changing as compared to your it will remain in the same place. Any idea what others configuration required? odom0: /diff_drive_controller/odom odom0_config: [false, false, false, false, false, false, true, true, false, false, false, true, false, false, false,] odom0_differential: false imu0: /imu/data imu0_config: [false, false, false, false, false, false, true, false, false, false, false, true, true, false, false,] imu0_differential: false

[@hoangminhtai4093]

hello sir, help me! Do the data sent from Ros to the arduino is the x and y coordinates right?

[@roboticsandroslearning8232]

Right motor and left motor velocities are sent to arduino

[@Simeon_Tsvetanov]

Thanks for the great content! I've sent you an email two weeks ago.

[@sathvickhk374]

Can you please upload the code for this

[@MemoxCid]

What data did you get from IMU?

[@ngocanhkieu5153]

Thanks for your video but i have a problem that when i run navigation using both amcl and ekf localization, there is an error in terminal: "Couldn't determine robot's pose asscociated with laser scan". How can you fix that?

[@MEan0207]

Can Holonomic robot use ekf package?

[@shouyudu936]

Hello Sir, Thank you very much for the video. Currently, I am encountering error with fusing the imu into robot_localization. I guess my question is what is the frame_id for your imu message, did you set it to base_link through a tf_broadcaster? or is it still in imu_link? Thank you very much!

[@gautamraj1513]

I can see that you are using imu_caliberated.py in the launch file. But the MPU 6050 driver package only contains imu_node.py. Can we get the src of imu_caliberated.py. Thank you.

[@gautamraj1513]

Hi can you please reply.

[@emiralpayar]

@@gautamraj1513 if you find I am also searching for that :(

[@VLSystems]

Thank you, exelent ))

[@VinayKumarTJ]

thank u so much..for info

[@johansebastian836]

In your imu code, what is the frame_id ?

[@sasieusp]

I am receiving could not obtain transform from imu_link to base_link warning. Are you getting the same?

[@anishdalvi]

Good Video. Can you tell the model name of LiDAR used by you? Is it RP LiDAR, if yes then which type? Also is this particular model of LiDAR reliable ?

[@roboticsandroslearning8232]

It is Rplidar A1 M8 and it is reliable

[@suleymankara1122]

Hello, thanks for sharing the video. We are trying to replicate your robot with your previous youtube video codes. Somehow i manage to run the robot. Unfortunately sensor fusion part is most difficult part of the ros. Especially tf of sensor fusion is question mark. I have alotof question but question answer part will take alot of time, inorder to that, can you please share your full ros code?

[@fernandoalgarin7017]

Hi, I'll be working on a similar project (Sensor fusion for pose estimation) and It would be nice if we can exchange ideas regard ROS programming , hope you read this!

[@ramanabotta6285]

As always 🤗.

[@ashokkumarchaudhary4723]

Nice demo. Can you suggest how to use MPU6050 or MPU9250 with Jetson Nano over I2C. Thanks in advance.

[@roboticsandroslearning8232]

Check out my previous video on mpu6050 kzbin.info/www/bejne/l17Ql3abrtCgj9k

[@siddhantpandey995]

I am doing same project in my final year can u provide the details of this project !

[@fernandoalgarin7017]

Hi I'm a young researcher from México , I would appreciate if we could stablish contact and share experience on the field! I'll be working in pose stimation and sensor fusion

[@ademcandemir4059]

hi sir i watched your both imu video and i have also have same mpu_6050 imu sensor and jetson nano your code is working the problem my plane in rviz is rotating about z axis very slowly it means it is not calibrated as i understood also sir i saw you add to your launch file imu_calibrated.py but you indicate first imu video us imu_node .py.can you share us updated package of autonomuos robot with mpu_6050 sensor ??

[@Ib.m86]

HI, I know it is very late :) , did you find anything ?

[@vipuldhamija8646]

Sir can you please tell us the valuable resources to learn ROS.

[@nhattran4833]

Hello friend, thanks for great video, can you explain about the different between EKF and ACML, what specific situation are they used in?

[@roboticsandroslearning8232]

Amcl is good when the wheel odometry is accurate and reliable. EFK is used to fuse multiple sensor data like wheel encoders, imu, gps, etc to generate accurate odometry even when one of the sensors are giving faulty data.

[@nhattran4833]

@@roboticsandroslearning8232 so we have to use ekf to estimate precise wheel odom, imu ... to create good odometry after this odometry will used by acml ?. Or we do not use EKF but use acml directly

[@Ruhgtfo]

Whoooool

[@TechTalks_with_Akshay]

Hi , how you make urdf ...for coustamize model

[@roboticsandroslearning8232]

I write it from scratch. But there is a way to convert solidworks model to urdf directly.

[@ArunKumar-sg6jf]

Where u learn this please reply for this ??

[@roboticsandroslearning8232]

Thanks to internet.

[@ArunKumar-sg6jf]

@@roboticsandroslearning8232 give any resources to learn

[@TNguyen-c6b]

Can you publish your source? Tks

[@Ananya_156]

Sir, do you have the source code or simulation for this? If so, please respond... 😊

[@fernandoalgarin7017]

Hi , I'll be working on a similar sensor fusion project on ROS and It would be nice if we can be in contact to share ideas , hope you see this!

[@skech_life1156]

Can i get ur information

[@prakharjain2548]

Quite an awesome demonstration. I have a couple of questions: 1. Is that a lidar module in the front of the bot? If so, what is the purpose of the it in this project? 2. Did you undertake this project out of your own interest or part of a project or under the guidance of some professor?

[@roboticsandroslearning8232]

1. Yes it is a Lidar. Lidar is to get the laser scan data which is used for localising the robot. 2. This project is out of my interest.

[@mouhamedali8038]

thank you sir for the video could you send the code on matlab of imu odometry data fusion

[@mirzasalem2766]

bro IMU is not work properly, can you kindly gave me som time?

[@roboticsandroslearning8232]

Use BNO005 IMU, I found it to be better than MPU's IMU

[@mirzasalem2766]

Can you kindly share your mail address? @@roboticsandroslearning8232

[@dineshlamarumba4557]

Can u share ur parameters?

[@TuNguyen-ev1uh]

can you publish your source code, bro!

[@TuNguyen-ev1uh]

How to calibrate it?