< back to news

First Haptics-1 dataset collection completed

Today, at 12:00 CET, NASA astronaut Butch Wilmore completed the data collection of the first full Haptics-1 experiment
Jan 8, 2015

NASA Astronaut Butch Wilmore completed our first full Haptics-1 data collection on-board ISS today. Now, all seven protocols of Haptics-1 have been recorded in both wall-mount and body-mount sessions! An exciting data-mining and analysis is comming up for us in the coming days and weeks. It will reveal in great detail which performances human crew can achieve with haptic force-feedback devices in space and which differences exist between the performance in space and on ground. In the wall-mount config. the joystick is fixed to the Rack's of the ISS and the astronaut stabilizes himself with the handrails (with the other hand) and with his feet. In the vest mount config. the joystick is fixed to the crew-member on his own chest. We believe that this force-closure on the human body might improve the then available hand-eye coordination of the crew member during the tests. Here, first, a brief summary of the recent events of the first Haptics-1 conduct on-board ISS along with a brief synopsis of the data we measure in the experiment.

30.12.2014 - 08:55 CET: First un-packing of the Haptics-1 Kit and start-up testing

08:55 CET: The first unpacking of the Haptics-1 Kit proved to be quite nerve racking. The unpack procedure started early, right after the crew post-sleep phase and took quite a bit longer than expected. Cool video footage of our Joystick floating in space freely during unpacking. Amazing to see that our hardware is now in orbit !

10:00 CET: Now Butch Wilmore had the system connected and tried firing it up for the first time. This is the moment. From ColCC we can only see the course video and can't follow the details. Crew takes the mic. and takes up comm's with ground. Butch Wilmore to ground control: "Switched the 1DOF on but the LED lit up only shortly then switched off again". "What shall I do?". This wasn't the news I wanted to hear. Turned out that the over current LED on the portable power supply went on. No good sign... 

10:20 CET: After warding off the immediate cancellation of our in-orbit session that was planned until 13:00, managed to secure the time allocated for a live debugging run. We have to get this going! It's space and you can't visit your hardware to debug even the most simple things. Communication with on-board crew is extremely difficult for us PI's. I speak with the user centre from ColCC to Denmark, via a dedicated voice link system. Denmark speaks with the US ground operators, since our candidate is a NASA crew member - via another voice-loop channel. US ground control speaks with the on-board crew - via yet another channel. 

around 11:00 CET: Multiple discussions with our user centre, myself, the flight directors and the NASA ground operators.. Butch is temporarily released to another task until we forward the detailed debugging plan.

around 12:00 CET: The cables connections double checked and problem re-occured. Seems that either our 1DOF Joystick has a short (would be very bad - but really can't imagine this can be the problem after our heavy testing and our careful integration of the hardware) or the outlet used on the portable power supply (PPS) is very sensitive to the current spikes produced during switch on (i.e. not the inrush - ofcourse we've measured that on ground) - or the spec that we've tested against wasn't sufficiently exact. To take it safe, we decide to use another PPS on-board unit and to change the switch-on sequence to first power-on the 1DOF and then to enable the PPS outlet. Last try! If this fails, we will have to change our power cable and include a more strict inrush filter.. this would cost a lot of time since it would entail a new hardware upload... it can't be a failure of our hardware since our device is built rock solid.

around 12:30 CET: Following the last debugging test, voice link from on-board crew brings the exciting and long awaited message: "It's on!" We are jumping up from our chairs! Released! Haptics-1 runs! I see in the video stream that the Tablet PC is on!

around 12:35 CET: We continue with the required software up-date procedure that was planned to install our latest version of the flight software. It's a more complex procedure difficult to monitor remotely with such indirect connectivity to crew. But after a couple of tries and a strict following through of the procedure our new flight software is installed successfully ! Version number confirmed through voice loop as read from the flight Tablet PC by crew.

around 13:15 CET: After another comm's interruption between ISS and Ground (happens approx. every 1-2 hours) we see the first video footage of how Butch operates the Haptics-1 protocol 2. In the meantime we've decided with flight directors and mission directors that we'd like to continue collecting data until our time is up, i.e. until the mid-day meal for Butch.. Luckily Butch continues until almost 14:00

13:50 CET: Butch got protocol 1 in the box along with some trials of protocols 2, 3 and 4. He finished protocol 5 entirely but now time was up, we leaves for his well deserved lunch break.. In the afternoon, after some other PR events and activities, our experiment will being stowed away for the next experiment slot planned for 05.01.


05.01.2014 - 08:55 CET: First full session of the Haptics-1 wall-mount experiment

It is the plan to fully complete the rack-mount session today.

08:55 CET: unpacking and installation of Haptics-1 run's very smooth and Butch finishes ahead of time by approx. 20 min.! Haptics-1 switches on spotlessly after a short procedure up-date concerning the switch-on sequence.

09:45 CET: Data collection starts! All protocols will be performed, except protocol 1, which is an automatic hardware identification. This one is only done once per crew.

Protocol 1: an automatic identification of all mechatronic components of the 1DOF Joystick. In one condition the joystick is locked and an identification signal excites the motor to measure the response of the output torque sensor on the handle-bar. In a second run the Joystick is free to move. Here the friction and inertial parameters are being identified. This measurement gives us all important information on the status of our hardware.

Protocol 2: a human-in-the-loop experiment where the joystick functions as a 'disturbance' to the human hand. Varying disturbance signals excite the Joystick to move in a random fashion and the human subject is asked to either simply hold the joystick, to resist the joystick or to comply with the joystick motion. This measurement delivers data which allows to infer the mass of the subject, the stiffness and damping settings of his entire upper extremity. This data allows us to model the human subject as a 'system' connected to robotic devices in space.

Protocol 3: This is the preferred tasks of the aircraft pilot astronauts. It's a tracking task in which a target has to be tracked with a cross-hair controlled by the joystick. Here, the joystick behaves like a positioning device of the cross-hair displayed on the tablet computer. The target signal is a random sinusoid. Positioning control bandwidth of moving the upper extremity is measured. We'll use this data to define the joint units of the later arm exoskeleton.

Protocol 4: Same as protocol 3 but now the force on the handle stick is used to control the cross-hair on the screen. THis measures that force controllable bandwidth of the crew in space. With this data we will know to which specifications haptic robotic controllers need to be built in space without requiring over-design or without loosing criticial information.

Protocol 5: A classical threshold test. We measure the minimum force differences a human test subject can discriminate in space at the level of the hand. Just noticeable differences (JND's) are measured for force/torque in a 2 alternative forced choice test (2AFC). This data gives us insight in how small difference of force/torque humans are capable to distinguish in space and thereby allows us to tailor the design of haptic controllers to the use-case in space. No need to build haptic devices that have higher resolution to what a human can possibly perceive.

Protocol 6: Same as 5 but now concentrated on stiffness differences. This data will be baseline data for a later experiment that will compare results with results obtainable through bilateral teleoperation over a variety of communication links

Protocol 7: A dedicated test to infer the optimum control parameters for a 'crisp' force feedback in space. The internal controller parameters of the force-feedback Joystick are varied and a 2AFC test is performed on 'crispness'. Improtant secondary data collection that will have a critical influence on our new exoskeleton design.

around 12:00 CET: Butch is finished ahead of time again! The mission planners are very happy! We've gained credibility of being a well functioning ISS experiment ! Now data is being transferred from the 1DOF system to ground for me to analyze in the coming days and weeks.. 


08.01.2014 - 08:55 CET: Completion of the Haptics-1 vest-mount experiment and thus the first full experiment run.

It is the plan to fully complete the vest-mount session today and thereby to complete the experiment with Butch Wilmore. Thank you commander Butch !

08:55 CET: unpack and install as smooth as before!

09:50 CET: start of the experiment data collection - all protocols !

12:00 CET: finished successfully ahead of time. All data in the box and data transferred to the SSC laptops for downlink to us. We've just completed the first complete run of the first haptics experiment that took place in space!

 

We'll be excited to report news once we get our next candidate Terry Virts to perform Haptics-1 as well. In total we'll collect data from 5 crew members in the coming months !


Category: Robotics Science