Schematic of Power management and flight control system of AST satellites.

AST production satellites will feature the same central propulsion unit as the test satellite Bluewalker 3, this will provide AST with information concerning flight heritage.

Both these satellite models share the new form factor invented by AST & Science.

Power management and flight control has a high level of redundancy, as can be seen in the schematic above filed by AST to the regulating body, the FCC.

We see multiple solar arrays and multiple batteries, each monitored.

Then there are two power conditioning units, PCDUs, two low voltage controllers, LVCs, feeding 2 x 5 flight computers, FCs.

One of these 10 flight computers then can control: Two propulsion units, PPUs, (which incorporates Orbion Hall Effect - Ion -Thrusters), four reaction wheels, RWs, for attitude control and more than 1000 highly distributed array elements, M.

>These more than 1000 array elements all have their own "separate power, processing, sensing and actuating".

From interpreting this quote I find it likely that the antenna array elements called microns also contain an magnetorquer each, and that the "system on a chip" being developed for the Bluebird microns not only controls communications, but also "power, processing, sensing and actuating", these micron systems are entirely software defined on the Bluewalker 3 test satellite, but will be cheaper SoCs on the Bluebirds.

Microns are since before known to be sandwiched containing an earth facing antenna array for cellular fronthaul covered by the antenna elements needed for highly directive beamforming, an solar panel that is facing away from earth, if not both ways / bifacial, an heat insulating layer between these elements, system on a chip SoCs (which are Application Specific Integrated Circuits containing processors)

Separate power may mean separate power cables or something like distributed battery pouch cells in each micron and if so a distributed energy storage.

Apart from giving the spacecraft extreme redundancy and highly efficient attitude control for changing ballistic coefficient between high and low drag configuration without using propellant, this amount of magnetorquers all individually controlled opens up the possibility of some interesting applications, such as in flight array shape control i

I'm an engineering student looking to explore career option in the industry. I saw a few companies that pique my interest in which they implement systematic control or automation at clients' sites to amplify productivity. The applications are so broad that I'm not sure what to look for.

For those who are in this type of industry and/or work as the engineers in this industry, I have questions for you. What are the valuable skills I would need to bring to get to work at those companies? What are some jargon that I should know when talking to the people in this industry? Are there areas I could improve or get an experience on while being a student?

Thank you in advance!

Hi!

I have a hard time finding people that have previously done some classes I'm interested in and had had previous bad experiences concerning the work load vs credits attributed or the course description attraction vs actual boring class.

Those are the classes I have my eyes on:

-Artificial neural networks

-Biological modeling of neural networks

-Distributed information systems

-Computational motor control

-Biomolecular structure and mechanics

If you have any insights/advices or just could tell your experiences I'd be happy to hear it!

Many thanks

Because I want your head on my branch.

Anyone knows a good course on distributed control of multi-agent systems?

When controlling motors, the commands are sent through communication IOs, not hard-wired. We have two sets of commands for starting the motor, START and STOP. From our side, the connections are as such:

DCS--Communication Card--Ethernet Cable--Ethernet Switch--Fiber Optic Cable--Fiber Optic Panel--Fiber Optic Panel--Fiber Optic--CCU--Ethernet Cable--MCC.

For the START command, 0 is non start and 1 is start, for STOP command, 0 is stop and 1 is non stop. For the following, I will use [0, 1] for non start and non stop for short.

Here is the current problem we are facing: when we start the motor, we go from [0, 1] to [1, 1]. No problem, the motor starts. Then we proceed to give [0, 1] and keep the motor running till future commands are sent. The problem is, when we would like to stop the motor and send [0, 0] to the Motor Control Center (MCC), they cannot receive this change and stay at [0, 1]. But I then try and give a conflicted command [1, 0] (usually this command cannot be sent unless you disable limitations), they receive the whole signal and sync to [1, 0].

What could be potential problems?

Hi all, I'm currently in the process of developing a fairly large event driven application that will communicate with an AB PLC, fanuc robots and various prebuilt systems provided by 3rd party vendors either via ethernet-ip, opc-ua or pure ascii messages over tcp/ip.

My application currently requests a schedule from a custom web application built by a 3rd party, provides the items in the schedule to a Allen Bradley PLC and tracks those items as they travel through different operations in a fairly large assembly line. As the items travel through the line, IO is triggered to provide different events in those operations and the current state is recorded. All of the items in the schedule are assigned unique ID's.

After some development with a colleague of mine that is programming the PLC, the subject of using a DCS has come up. I'm not familiar with any of the DCS solutions provided by the major vendors apart from Rockwell's PlantPAX, which I have minimal experience with but feel wouldn't meet the requirements of the project.

Has anyone here used a DCS to communicate with a restful API that responds with json data or are there any DCS systems capable of doing so? Would there be any time saving benefits of choosing a DCS system? I'm currently using a few 3rd party libraries to handle communicating with ethernet-ip and opc-ua devices.

Additionally I'll also need to read files stored on the local file system that are formatted in json to provide part details nested in a collection of jobs.

Any shared experience is greatly appreciated.

Hi;

This is for those who like Darkest Dungeon and also like programming. I created a Git wrapper that adds some of the Narrator's quotes to Git. If you are curious, access the link bellow and turn your volume up.

https://github.com/prvst/darkest-git

cheers.

Bungie has made many improvements to armor, post Armor 2.0. However, managing armor in Destiny 2 is still a cumbersome, RNG dependent, and grindy process that gives very little agency and control to players.

The stat totals for armor are great, but I have four ideas to improve the armor grind toward a more-RPG-like player driven experience:

1. Simplify stat rolls to multiples of 5 and add minor bonuses to mid-tiers
2. Add set bonuses, improve masterworking, and reduce the cost of affinity switching
3. Allow players to redistribute some stat points at a vendor/workbench.
4. Allow us to craft our own armor from armor blueprints.

Improving the Armor Drop Experience:

The hunt for good armor can be pretty demoralizing. Basically, base stats only benefit you at multiples of 10; yet armor drops with stats at increments of 1-2. This ensures that you always have wasted points and inflates the grind. The armor system used to be simpler, but I'd love to see it evolve in a way that adds some depth:

• Stats should only roll in multiples of 5. Most stats categories would have a min of 5 (up from 2) and max of 20 for legendaries/25 for exotics.
  • Max base stat total for legendary armor piece remains at 70.
  • Exotics max base stat total should be 75
• Class armor items should roll with +5 in one stat category.
• Stats between tiers (5, 15, 25, and so on) should offer a minor benefit (25% of the normal tier bonus?).
• Set-bonuses could be added to future activity-based legendary armor. Tiered bonuses that trigger when wearing 1, 2, or 3 pieces of an armor set.
  • Bonuses could be minor &/or activity-specific for balance purposes.
  • Example: Crystocrene set could offer damage resistance against Fallen while on Europa (5/10/15% for 1/2/3 pieces).
  • Example 2: Daito armor (prophecy Dungeon) could grant bonus stability/handling/reload speed while using Ikelos weapons for 1/2/3 pieces.

Other quality of life changes:

• Masterwork bonus could be increased to +15 from +12 points.
  • players could be allowed to (re)allocate the +3 points to one stat of their choosing.
• The cost of switching elemental affinities should be no more than legendary shards and glimmer. Stop punishing people for wanting to try out builds.

Introducing set-bonuses and reducing conflation should allow players to truly invest in, yearn for, and enjoy multiple armor offerings. Next, let's look at increasing player agency and investment in armor customization.

Hi All, I'm taking a classic graduate course from MIT https://pdos.csail.mit.edu/6.824/ on my own (I'm not enrolled, just doing the course independently). There are 4 different labs in Go lang I plan to redo in Scala (MapReduce, Raft, Fault-tolerant Key/Value Service, Sharded Key/Value Service). There is also a bunch of lectures available on YouTube: https://pdos.csail.mit.edu/6.824/schedule.html. I would like to know if there is any interest in forming a study group to motivate each other to complete this course and work on the labs together.