Hello all,

After lots of work I have finaly my robot driving control with a joystick from my desktop. I have created an simple URDF of the robot which has 3 ultrasone sensors. This data I receive on RVIZ. But the next step is using odometry and see the bot moving in RVIZ. I wrote a script that computes the wheel velocities into linear and angular velocities and publish them as a nav_msgs.msg /Odometry and named the topic /odom.

When I drive the robot and do a rostopic ech /odom I receive the data a screenshot below: So this seems to work, I guess. But when I try to visualize it with RVIZ it tells me that there is no odom link to transform. The message I saw was:

Transform [sender=unknown_publisher]

I have also done a rosrun tf view_frames this showed me the following:

Here it is clear that there is no /odom frame. So how could I fix this? Do I need to change my URDF file or do I make a mistake in RVIZ I don't know where to look at the moment anymore.

I hope someone can guide me a bit

Edit1: (Snippet of first attempt using odom from wheel encoders

    def calculate_odometry(self, right_vel_rpm, left_vel_rpm):
        wheelbase = 0.205                                                                                   # Wheelbase in m
        wheel_radius = 0.036                                                                                # Radius of wheel in m

        right_vel_mps = (2 * math.pi * wheel_radius * right_vel_rpm) / 60.0                                 # Convert rpm to m/s right side
        left_vel_mps = (2 * math.pi * wheel_radius * left_vel_rpm) / 60                                     # Convert rpm to m/s left side

        linear_velocity = (right_vel_mps + left_vel_mps) * 0.5                                              # Compute the linear velocity
        angular_velocity = (right_vel_mps - left_vel_mps) / wheelbase                                       # Compute angular velocity in rad/s

        current_time = rospy.Time.now()
        dt = (current_time - self.last_time).to_sec()                                                       # Compute integration time and covert from us to sec

        self.x += linear_velocity * math.cos(self.yaw) * dt
        self.y += linear_velocity * math.sin(self.yaw) * dt
        self.yaw += angular_velocity * dt

        self.publish_odom()

        self.last_time = current_time

def right_vel_callback(self, msg):
      self.right_vel_rpm = msg.data
      self.calculate_odometry(self.right_vel_rpm, self.left_vel_rpm)

def left_vel_callback(self, msg):
      self.left_vel_rpm = msg.data
      self.calculate_odometry(self.right_vel_rpm, self.left_vel_rpm)

def publish_odom(self):
      odom = Odometry()
      odom.header.stamp = rospy.Time.now()
      odom.header.frame_id = 'odom'
      odom.child_frame_id = 'base_link'

    # Populate position
      odom.pose.pose.position.x = self.x
      odom.pose.pose.position.y = self.y
      odom.pose.pose.position.z = 0.0
      odom.pose.pose.orientation.z = math.sin(self.yaw / 2.0)
      odom.pose.pose.orientation.w = math.cos(self.yaw / 2.0)

    # Populate velocities
      odom.twist.twist.linear.x = 0.0
      odom.twist.twist.linear.y = 0.0
      odom.twist.twist.angular.z = 0.0

      self.pub_odom.publish(odom)

https://wiki.ros.org/navigation/Tutorials/RobotSetup/Odom

geometry_msgs::Quaternion quat = tf::createQuaternionMsgFromYaw(odom_yaw);
tf_msg.header.seq++;
tf_msg.header.stamp = ros::Time::now();
tf_msg.transform.translation.x = odom_x;
tf_msg.transform.translation.y = odom_y;
tf_msg.transform.translation.z = 0.0;
tf_msg.transform.rotation = quat;
odom_broadcaster.sendTransform(tf_msg);