Ask Your Question
0

Robot localization:using 3 sensors(wheel encoder,imu,gps) in curves we are getting drifts in output

asked 2018-11-14 07:28:31 -0600

moulya gravatar image

updated 2018-11-27 22:49:47 -0600

Hi Everyone,

I am using robot localization package to fuse 1 wheel encoder,1 imu and 1 gps.

  1. For doing localization , I am using 1 state estimation node(ukf) inputs: encoder(/odom/data),imu(imu/data), gps(odometry/gps). output: odometry/filtered
  2. navsat_transform_node inputs:gps(/fix1) output: odometry/gps

I am able to get output for this configuration but the problem is whenever i am running these nodes in curverd path it is showing drift.

In straight path ,the output is as expected.

  1. I want to know how to solve this issue in curved path and at the end.
  2. I want to know why static transform is required , i m giving yaw(starting heading of robot) manually here but it ll change
    everytime what can i do to do that automatic.
    1. Please view the launch and yaml file and suggest us what can we do to correct the issue.

edited-

The input messages for sensors used are,

odometry data(topic:/odom/data)

header: 
  seq: 18
  stamp: 
    secs: 1543233105
    nsecs: 933596449
  frame_id: "odom"
child_frame_id: "base_link"
pose: 
  pose: 
    position: 
      x: 0.0
      y: 0.0
      z: 0.0
    orientation: 
      x: 0.0
      y: 0.0
      z: 0.0
      w: 0.0
  covariance: [0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03]
twist: 
  twist: 
    linear: 
      x: 0.0147113889434
      y: 0.0
      z: 0.0
    angular: 
      x: 0.0
      y: 0.0
      z: -0.00374101494052
  covariance: [0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.03]

IMU data(topic:/imu/data)

header: 
  seq: 313
  stamp: 
    secs: 21
    nsecs: 484000000
  frame_id: "imu_link"
orientation: 
  x: 5.24012097576e-05
  y: 0.000117985712522
  z: -0.988258955695
  w: 0.152788153412
orientation_covariance: [0.001225, 0.0, 0.0, 0.0, 0.001225, 0.0, 0.0, 0.0, 0.001225]
angular_velocity: 
  x: 0.0
  y: 0.0
  z: 0.0
angular_velocity_covariance: [0.0004, 0.0, 0.0, 0.0, 0.0004, 0.0, 0.0, 0.0, 0.0004]
linear_acceleration: 
  x: 0.0310470898156
  y: -0.00114743773105
  z: 9.7057897167
linear_acceleration_covariance: [0.009604, 0.0, 0.0, 0.0, 0.009604, 0.0, 0.0, 0.0, 0.009604]

GPS data(topic:/fix1)

header: 
  seq: 4
  stamp: 
    secs: 23
    nsecs:   4000000
  frame_id: "gps"
status: 
  status: 1
  service: 1
latitude: 49.8548095709
longitude: 8.59120397786
altitude: 0.92988040448 ...
(more)
edit retag flag offensive close merge delete

Comments

1

Please update your question with the following:

  1. Show an image of what it is you're seeing
  2. Provide a sample input message for every sensor input
Tom Moore gravatar imageTom Moore ( 2018-11-26 04:03:04 -0600 )edit
1

Thanks @Tom Moore for reviewing the question.

  1. I am not able to upload images because it's asking me to acquire some points.
  2. I have updated the question with sample input messages for the sensors.
moulya gravatar imagemoulya ( 2018-11-26 06:10:50 -0600 )edit
1

I gave you some karma. Care to update the question with images?

Tom Moore gravatar imageTom Moore ( 2018-11-27 07:38:34 -0600 )edit

Thanks @Tom Moore for giving me karma .I have uploaded the images.

moulya gravatar imagemoulya ( 2018-11-27 22:45:34 -0600 )edit

1 Answer

Sort by ยป oldest newest most voted
1

answered 2019-01-09 07:34:48 -0600

Tom Moore gravatar image

Just so I understand, you are concerned that the paths aren't right on top of each other? Are you visualizing in top down ortho mode in rviz?

image description

Check the Z coordinates of the EKF output and the navsat_transform data. And are you sure the straight-line segments line up? It's a line, so it's impossible to say from your image whether the poses are offset from each other.

And what static transform are you not certain of?

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

3 followers

Stats

Asked: 2018-11-14 07:20:06 -0600

Seen: 630 times

Last updated: Jan 09 '19