The motor driver part of the robot can be turned on using the following tutorial:
https://trac.vicos.si/ros/wiki/ATRVMiniSetup
Next up, press the green button on the rear of the robot (or the one inside) to turn on the computer.
After that it's set up to connect to ViCoS_5Ghz under the following static IP:
SSH:
ssh atrv@192.168.33.251
The username and password are set to 'atrv' and 'visionm4n'.
Once inside you can start all of the currently implemneted drivers by running:
roslaunch atrv_mini_ros atrv_full.launch
The functionalities launched will be:
-
atrv motor drivers
-
flir ptu driver
-
realsense2 camera
-
led controller
-
soundplay node
The idle movement demo can be launched using (note - also launches flir_ptu_driver):
roslaunch face_led idle_pantilt.launch
Test driving the robot can be done using:
roslaunch atrv_mini_ros teleop.launch
If it's running first turn of the PC by either running something like:
sudo poweroff
or holding the green button inside the robot if a wifi/ssh connection cannot be established (acts as a normal shutdown button, the red one is a reboot button).
Then turn of the motor driver part by accessing the menu on the rear display of the robot, entering the [PWR] submenu and then pressing [Kill PWR].
The used pan tilt ROS driver is http://wiki.ros.org/flir_ptu_driver, with the docs on the linked wiki page.
The pan tilt can be controlled through the JointState topic /ptu/cmd, where angles can be set in radians.
There's a helper script that can make it easier to test:
rosrun flir_ptu_driver cmd_angles 0.9 0.3
The first value being the pan and the second the tilt. Make sure to set the values in a reasonable way to avoid collisions.
The package face_led contains a demo script and a topic controlled script that interfaces with the arudino in the head that sets the actual LEDs.
One can run them manually like so:
rosrun face_led led_controller.py
python demo.py
The led controller subscribes to three topics:
/face_emotion (String)
/face_eye_anim (Float32MultiArray)
/face_talk_anim (Float32)
First one takes a string that sets the overall face disposition our of the following: ["blank", "neutral", "neutral_left", "neutral_right", "neutral_happy", "neutral_frown", "furious", "mad", "stupid", "sad", "miserable", "happy", "nauseated", "surprised", "loading"].
The second one takes a float that sets how long the mouth should do a talking animation from the moment the message is received. The animation will be applied to the currently set emotion set by the topic above, as will the eye_anim topic, which takes a float array of two variables (left and right eye) that specifies how many seconds each eye should blink (speed of the blink itself). If -1 is sent, the specified eye will not blink.
An idle head movement animation that includes blinking and changing emotions is implemented in idle_anim.py.
The Realsense depth camera requires the Realsense SDK https://github.com/IntelRealSense/librealsense and the ROS driver https://github.com/IntelRealSense/realsense-ros which are both installed on the robot.
They can be launched using one of the launch files in the realsense2_camera package.
Related topics:
/camera/color/camera_info
/camera/color/image_raw
/camera/color/image_raw/compressed
/camera/color/image_raw/compressed/parameter_descriptions
/camera/color/image_raw/compressed/parameter_updates
/camera/color/image_raw/compressedDepth
/camera/color/image_raw/compressedDepth/parameter_descriptions
/camera/color/image_raw/compressedDepth/parameter_updates
/camera/color/image_raw/theora
/camera/color/image_raw/theora/parameter_descriptions
/camera/color/image_raw/theora/parameter_updates
/camera/depth/camera_info
/camera/depth/image_rect_raw
/camera/depth/image_rect_raw/compressed
/camera/depth/image_rect_raw/compressed/parameter_descriptions
/camera/depth/image_rect_raw/compressed/parameter_updates
/camera/depth/image_rect_raw/compressedDepth
/camera/depth/image_rect_raw/compressedDepth/parameter_descriptions
/camera/depth/image_rect_raw/compressedDepth/parameter_updates
/camera/depth/image_rect_raw/theora
/camera/depth/image_rect_raw/theora/parameter_descriptions
/camera/depth/image_rect_raw/theora/parameter_updates
/camera/extrinsics/depth_to_color
/camera/extrinsics/depth_to_infra1
/camera/extrinsics/depth_to_infra2
/camera/infra1/camera_info
/camera/infra1/image_rect_raw
/camera/infra1/image_rect_raw/compressed
/camera/infra1/image_rect_raw/compressed/parameter_descriptions
/camera/infra1/image_rect_raw/compressed/parameter_updates`
/camera/infra1/image_rect_raw/compressedDepth
/camera/infra1/image_rect_raw/compressedDepth/parameter_descriptions
/camera/infra1/image_rect_raw/compressedDepth/parameter_updates
/camera/infra1/image_rect_raw/theora
/camera/infra1/image_rect_raw/theora/parameter_descriptions
/camera/infra1/image_rect_raw/theora/parameter_updates
/camera/infra2/camera_info
/camera/infra2/image_rect_raw
/camera/infra2/image_rect_raw/compressed
/camera/infra2/image_rect_raw/compressed/parameter_descriptions
/camera/infra2/image_rect_raw/compressed/parameter_updates
/camera/infra2/image_rect_raw/compressedDepth
/camera/infra2/image_rect_raw/compressedDepth/parameter_descriptions
/camera/infra2/image_rect_raw/compressedDepth/parameter_updates
/camera/infra2/image_rect_raw/theora
/camera/infra2/image_rect_raw/theora/parameter_descriptions
/camera/infra2/image_rect_raw/theora/parameter_updates
/camera/realsense2_camera_manager/bond
/camera/rgb_camera/auto_exposure_roi/parameter_descriptions
/camera/rgb_camera/auto_exposure_roi/parameter_updates
/camera/rgb_camera/parameter_descriptions
/camera/rgb_camera/parameter_updates
/camera/stereo_module/auto_exposure_roi/parameter_descriptions
/camera/stereo_module/auto_exposure_roi/parameter_updates
/camera/stereo_module/parameter_descriptions
/camera/stereo_module/parameter_updates
The package robobrain should contain all robot behaviour related nodes.
The dnn_detect package can be launched and remapped to the realsense topics as follows:
<node pkg="dnn_detect" name="dnn_detect" type="dnn_detect" output="log" respawn="false">
<param name="image_transport" value="compressed"/>
<param name="publish_images" value="true" />
<param name="data_dir" value="$(find dnn_detect)/model"/>
<param name="class_names" value="background,aeroplane,bicycle,bird,boat,bottle,bus,car,cat,chair,cow,diningtable,dog,horse,motorbike,person,pottedplant,sheep,sofa,train,tvmonitor" />
A wip test can be found in atrv_debug.launch.
The speakers are currently connected through the analog audio jack but an alternative should be found to avoid induction noise.
Speaker volume can be adjusted using the following command:
sudo amixer cset numid=1 80%
or by using the 'alsamixer' interactive setting board.
The robot contains two microphones in its 'ears' that are connected to a 'SAMSON Go Mic Connect' stereo microphone.
There's no ROS integration at this time.