Journalist Report – April 18th

Crew 297 – Janus 1 Journalist – David Laude

Janus 1’s first double EVA on Sol 4, taken on another beautiful Martian day by two teams of two. One team manned by both Matts worked on burying the nuclear reactor in regolith, while at the same time Pawel and Dave got practice with the robot system provided by Nicholas Conlon of CU Boulder.

The two Matts took turns digging until after 54 minutes of toil they had a hole big enough to hold the entire mock reactor. A few issues came up along the way including Matt’s (does it matter which one?) communications head piece that slid of his head until repositioned. That’s part of the learning in sim experience. Afterwards, they removed the reactor and covered up the hole. We will hear more about the reactor in future reports.

Meanwhile Pawel and Dave set up the mobile robot, named "Case", after a robot from sci-fi movie "Interstellar". The process was long, due to our not having done it before, but it worked the first time. Pawel did the driving from a laptop while Dave walked along with his new friend Case, while watching for obstacles. Case can be programmed from the laptop to drive a particular course, but sometimes Pawel had to manually drive it from his laptop computer.

The main goal of the study with Case is to understand how future astronauts, in current real-world robotic applications, where users rely heavily on telemetry, map data, and intuition in order to infer how competent a robot will be in a given environment. Telemetry can consist of a variety of data, however in our experiments, telemetry will include the robot’s position, heading, velocity, battery level, and other state information. Map data consists of a displayed map with iconography indicating features such as positions of the robot, waypoints, hazards, and other relevant information. This information, while valuable, can be confusing for non-expert users whose mental model of the robot’s competence is incomplete or inaccurate, tedious to follow and monitor, and can lead to poor human decision-making. Instead, this research focuses on developing more human-centered approaches to convey robot competency.

In an unusually productive day, Sean and Matt S. also gave the rest of the crew an introduction to their software based "Toolset for Shared and Long-term Document Management and IT Operations". More on this in a later journalist report.

Today’s person of interest is Executive Officer Matthew Storch. He has B.S. and M.S. degrees in electrical engineering from Stevens Institute of Technology and a Ph.D. in computer science from UIUC. He has worked as a software engineer and has held various engineering management positions for 35 years He is currently acting VP Engineering and CTO of a small (30 person) company that is building a specialized physical infrastructure management product (target audience is large corporations and government institutions). Outside of work, Matthew has a long-standing passion for adventure, technical achievement and unusual experiences that has led him to becoming an airplane pilot, a gyroplane pilot, sailboat & powerboat operator, and a submersible pilot.

Copyright © The Mars Society. All rights reserved. | Main Site