I can do the same with almost any IP camera and even USB one with low latency and maximum quality the camera can supports. If anyone is interested in such setup I can do a quick write up about or maybe a repo with the configs needed.
I want to do same thing with smart plugs. There's not a single smart plug on market that won't connect to cloud to control plug from anywhere. I don't want to control from anywhere dammit. I just want control on my LAN.
If starting anew Thread is probably a better bet. Such as the Eve smart plug.
But with both zigbee/thread the problem is shifted towards the gateway instead so you'll have to do it yourself / pick one tu trust there. But here I think/hope Thread is going to have better interoperability.
And if you don’t want to deal with the hub potentially being your next issue, you can always throw home assistant on a computer / raspberry pi / whatever and control them locally. If you want remote access when you’re off your network, it’s up to you to figure out how to securely get the bits from your phone to your home assistant server.
+1 for Cloudfree. I don't know much about IoT but I bought 5 of their plugs and they literally took under a minute to get set up and running. Another few minutes to get data off of them and into Influx Cloud using Python.
I wish more manufacturers provided this sort of thing!
I'd recommend looking at Tasmota and/or ESPHome support list and getting one of those. Both are 100% open source and do not talk to any external servers. Some devices can be reflashed over-the-air and some need to be taken apart and flashed via a serial header. A few devices, such as from Athom, come preloaded with Tasmota. Tasmota has a list of IN-specific devices: https://templates.blakadder.com/in.html Here is an example of opening one up and connecting to the serial header: https://templates.blakadder.com/oakter_oak_plug_plus_16A.htm...
For the light switch in my room I slapped together a relay and a raspberry pi pico and a Bluetooth board (I think I saw recently that the pico w now has working Bluetooth as well?)
Now I can control the lights from my bed even when the LAN is down.
As others have said, look for Zigbee devices. I have an Osram SomethingOrOther, that connects to my Hue bridge, which isn't connected to the internet. Philips makes a similar plug.
The fact is that the Hue bridge supports connecting to a cloud for control from anywhere, so they'll advertise that since it's what many people are after. But it works perfectly fine offline, and you can manage it via Home Assistant, for example.
Shelly is pretty good for this. Their Android app really sucks badly but the good thing is you only need it for the initial setup.
Whatever you do, don't buy TP-Link. Those had local control but they removed it due to "security concerns" in an automatic update. And of course promoting their data mining cloud had nothing to do with it.
Other than every single one that uses zigbee/zwave and not Wi-Fi. You can also use services like Cloud cutter or select devices listed on the Blakader website of tasmota-flashable silicon. Or if you're willing to go a little deeper into the weeds, you can use ESP home for more fine-grained control.
I was gifted a new rando chinese IP cam that uses the mobile app. It's not googlable. I've been using as a outside security camera. The camera saves to an SD card continuously, and has a paid option to use the cloud which I did not take. It has a mobile app, KP WIFICAM, which lets me view the feed in real-time even over the Internet.
I can even view or download the past few days of recordings on the SD card from a menu. However, they are separated into chunks of 10 minutes, so I would have to manually download nearly 100 files files (tap, Download, wait to finish, move to next clip) to get a single day of recordings.
How could I automate/simplify the downloading of these clips from the SD cards to my laptop? How is the mobile app listing the recordings and fetching them?
I ran a tool called Agent-DVR and it found the camera feed using a format called ONVIF. So I think the mobile app uses this protocol https://stackoverflow.com/questions/29159422/get-data-from-l... , is there a tool like this available for Linux I could use without spending a couple of days installing a developer environment and writing my own code?
One simple workaround might be to get a wireless SD card? There are a few of these for DSLRs, where they have some (relatively small) amount of storage but come with the ability to join the network and share it via samba seamlessly...
If you have a device on which you can run my app (streamie.co), it’ll record your camera 24/7 to a NAS or S3 and write out .mov files (also ten minute segments), but they’d be trivially combined with ffmpeg. As long as you don’t need more than three cameras, it’s all free.
you should sell an application to manage many cams from a pc with a phone app to access it remotely.
The significant other has 4 different apps for cams. They each have silly issues, sd cards are to small and they all charge wild fees for cloud storage that they didn't disclose before purchase.
I can buy new ones but with the cheaper ones it is more likely a repetition of the mistake made 4 times already.
The software can be 6 times as expensive as the cams and still be entirely worth it.
What I also want is someone in a low wage country watching a wall full of cams cycling from one cam to the next. That service western security companies charge thousands per month for. Robbery right there!
I've got some Reolink Argus 2s, which are probably Linux under the hood. They're not set up to stream locally without an app, unlike their wired models, but there are likely functionality limitations due to being battery powered.
This would be great. Google sent out an email not long ago that they were discontinuing nest cams. Mine still works great but soon will become a paperweight, otherwise.
Personally, not just twitter, all social media I’m happy to see it dies, and no, “stop using it” mindset isn’t an argument when it severely affects the majority of the people around you, and start not just influencing the populace opinions but shaping the next generation. Who ever controls that narrative in the media get to decide what’s the next laws to be considered, after all, that’s how democracy works.
Exactly this, I’m happy of what’s happening with twitter and Reddit and the likes, maybe it’s the reality slap the users needed to get back to smaller communities.
A lot of generic fluff, nothing is specific.. what was the frequency band/s they used? Why line of sight is needed for such comms when it can be done otherwise? Do the chopper needs to be connected all the time for command and control channels, or are there any autopilot? Or is the connection meant is the payload one (camera etc.), I assume there’s or it would’ve crashed in landings if signal was lost and autopilot landed in an unsuitable location? Why move the rover at the same time while the chopper was flying? Shouldn’t be stationary for the best signal? Especially when there’s no gnss system there, why add that complexity of moving two of them (chopper/rover)? And if only one was moving that time, how they missed the fact they are going behind a hill? Was the battery dead when they lost signal? How they charged it after losing it? How they found the chopper in the first place? Did they find it by ground search and then flew the chopper back to the rover? How the battery survived that time?
I don’t know there are a lot of other questions I have in mind, that article barely explains anything..
I think the reason they don't explain much is because a site like space.com would have covered all that many times over by now. Some answers:
1. The copter isn't able to communicate with satellites or earth directly, so it needs to communicate with the rover itself. This is harder without line of site.
2. The copter is all autonomous. Routes are planned and sent for the copter to autonomously fly, because direct control would be infeasible with the latency on Mars.
3. The copter is charged with solar power, and needs time to charge between flights.
4. The main rover has other missions that are independent of the status of the copter. The copter is well well well past the original mission goals, so if it gets left behind that's a shame, but the main rover's missions are way more important.
>The copter isn't able to communicate with satellites or earth directly, so it needs to communicate with the rover itself. This is harder without line of site.
Obviously no satcom on there, but even when the chopper communicates with the rover, a LoS isn’t needed, different comms bands can survive that.
>The copter is all autonomous. Routes are planned and sent for the copter to autonomously fly, because direct control would be infeasible with the latency on Mars.
Great, so if it’s planned how they end up with that situation? Was it a calculation errors? Or they suddenly decided to move the rover and lost comms?
>The copter is charged with solar power, and needs time to charge between flights.
Ok now this is interesting, got any more details about it? As the size of the panels would be big and might hinder the chopper movement, unless something innovative was done.
>The main rover has other missions that are independent of the status of the copter. The copter is well well well past the original mission goals, so if it gets left behind that's a shame, but the main rover's missions are way more important.
Ok that’s good to know, still curious about the whole process as I do have a current project with interoperability between UAV/UGV so any previous lessons learned would be great!
