Feature Request Updates: What’s Gaining the Most Attention!

Hello everyone,

We’ve received a lot of feature requests here on the forum, and we’ve noticed that many of you have expressed similar interests in some of the same ideas. As a result, we’d like to share updates on some most popular feature requests, including: map editing optimization (split area and straight line), tow mode, smart vision, schedule feature enhancements, route optimization and slope algorithm optimization.

Here’s an overview of the current status and upcoming plans:

1. Map Editing Optimization

We’re actively working on a comprehensive map editing optimization program. Two key features being included are the ability to split areas and draw straight lines during mapping.

  • Area Splitting: After splitting an area, you will be able to retain all sections of the area.

  • Straight Line Feature: When mapping, you’ll be able to create a straight line between two points, even if the actual path your Yarbo takes isn’t a straight line.

We plan to release these two map editing improvements as part of the complete optimization program. However, since we’re still working on other features within the program, the release will be delayed to June/July.

2. Tow Mode

The first version of the tow mode is expected to be included in the next major update (pending internal testing and approval). This mode has been designed with attachments in mind, allowing you to adjust the path spacing to prevent jack-knifing.

Additionally, the reverse function will be disabled while using tow mode to minimize potential risks. Initially, tow mode will be available for the SAM module.

3. Smart Vision (Camera Feed While Yarbo is Operating)

We’re currently testing the smart vision feature optimization with PPP users and are collecting feedback. During our internal testing, we’ve found that smart vision occupies some of the system’s resources, which can occasionally cause frame drops. This happens because the system prioritizes Yarbo’s operational tasks over the camera feed.

Once we’ve gathered enough feedback, we’ll continue optimizing the feature. The expected release date for smart vision is now set for July/August.

4. Schedule Feature Enhancements

Based on user feedback, we’re working on two major improvements to the schedule feature:

  • Automatic Resuming unfinished work plan during schedule: Currently, when the end time reach, Yarbo lawn mower will stop the wok plan and automatically recharge. Some users have reflected that they want Yarbo to automatically resume the work plan the next day. This feature is planned for release in the next two major updates.

  • Interval-Based Scheduling: We’re exploring the possibility of allowing you to schedule mowing at regular intervals, such as every three days. While this idea was recently raised, we’re still in the early stages of planning. We don’t have an estimated release time for this feature yet.

5. Route Optimization

  • Pathway Optimization

To reduce lawn wear caused by Yarbo repeatedly using the same pathway, we’re working on a pathway offset feature. This will allow Yarbo to shift to the left or right side of the original pathway when traveling between areas or between the docking station and the area. With this offset, Yarbo will alternate between the left, original, and right pathways, changing the route every three passes based on our current design.

This feature is still under development, and we plan to release it alongside the map editing optimizations in June/July.

We’re also considering an option to allow you to create multiple pathways, with the system automatically selecting different routes during travel. This feature is still under evaluation.

  • Route Optimization Between Areas or Docking Station

Currently, Yarbo sometimes cuts through already mowed areas when traveling to another zone, which can affect the tidiness of the lawn. To address this, our solution is to have Yarbo travel along the perimeter of the area, avoiding previously mowed sections, though we understand this may slightly reduce the efficiency of route planning.

We have scheduled the feature for the third upcoming release.

  • Turning Before Docking Station Causing Turf Damage

To address the issue of Yarbo potentially causing turf damage when turning before reaching the docking station, brysach2 has provided two possible solutions, which we have discussed with the product team to make an initial decision:

  1. Three-Point Turn Option: After evaluation, we decided not to proceed with this option, as performing a three-point turn on the pathway could increase the risk of Yarbo going out of bounds.

  2. Multiple Turning Points: We are considering the option of allowing users to set multiple turning points along the pathway, enabling Yarbo to turn at different locations. Development and testing will begin when more development resources become available, as we are currently focused on other priorities.

6. Optimization of Mowing Performance on the Slope

Some users have observed uncut patches of grass on slopes. This issue is primarily caused by two factors:

  1. Route alignment relative to the slope: When the mowing route is oriented roughly perpendicular to the slope (as shown in the illustration), gravity causes the machine to slip slightly downhill, leading it to deviate from its planned trajectory. Although the system will eventually correct the path, this deviation results in curved or incomplete mowing lines, leaving behind strips of uncut grass.

  1. Attitude (tilt) of the mower affecting positioning accuracy: On slopes, the robot experiences continuous changes in pitch and roll angles. These changes can shift the physical position of the GNSS antenna, especially its height and orientation. As a result, the reported RTK position might not precisely match the actual cutting edge location. This discrepancy becomes more pronounced on steeper slopes or uneven terrain, and can lead to slight mismatches between the planned route and the actual mowing path, contributing to uncut areas.

We are planning a project to optimize mowing performance on slopes. However, our development resources are currently focused on addressing the issue of the machine going out of bounds. As a result, this project is not our top priority at the moment, and we have not yet developed a fully mature proposal. Our initial design considers the slope pattern proposed earlier, as well as improvements to the IMU system to enhance performance. We may have additional proposals once our development resources are allocated to this program.

We appreciate all the feedback and suggestions you’ve shared. While some features are still in development, we’re committed to continuous improvements and will keep you updated on our progress. Stay tuned for more updates, and thank you for your support!

8 Likes

Awesome update!

Quick point: Reversing while towing is needed to do things like back a lawn sweeper into its dump spot, so we should still be able to go to manual control to do that, then maybe resume the towing work plan. Or something like that. Yes?

3 Likes

Thank you for the update! Good progress!
So no May update planned anymore? Everything pushed out to June /July?

I do understand that your development resources are limited, but making the optimization of the slope algorithm not priority is very sad to hear, 80% of my yard is on a slope, Yarbo mows the flat areas very nicely (600m²), but the steep slopes are a completly different story. You guys advertised Yarbo as a mean mowing mashine for steep slopes and rough terrain, at the moment all it does, is drawing precise lines with grass it misses every turn.

For me, having almost nothing but steep slopes and judging by the number of threads on this forum regarding mowing on slopes, the optimization of the mowing algorithms should be your top priority!

5 Likes

OK, here’s another Ken’s Questionable Hack!

Your mowing on the slope does work, but it leaves unmowed stripes, correct? Let’s assume Yarbo is going to fix this soon, so you want something just for now.

  • Not automated, but low effort: If you mow once and then change the Boundary Set Back and mow again, does it shift the rover over enough to get those stripes? Each time you mow you could vary your mowing lines using Boundary Set Back and at least put a band-aid on this. Of course it’s not automated, but if it works “for now,” will it help? (I’m assuming you want to keep the same mowing angle because it works best with the slope – at least this may shift the rover over a little.)

  • Automated: If the above does work, what if you map an intentionally overlapped Area and set the Boundary Set Back different OR mapped the overlapping Area with its boundaries slightly different to shift the mowing lines over? You could mow the original Area on one day and the new temporary Area the next time – automated. Or if you just created a small temporary Area overlapping the stripes and scheduled that. No? Yes?

2 Likes

Interesting idea, I mow in diamond pattern. I recently changed perimeters from 3 to 2, that moved yarbo over a little, but not enough to compensate a 20cm (~10in) strip. What helps for now is mowing frequently in different directions.
After 2 or 3 weeks it’s OK, it repeats when I can’t mow because of bad weather.

2 Likes

Varying the perimeter passes is a good way to catch some of the missed and if you are having wear at the turns to vary where it turns.

I think this is the problem with the slopes (or at least in addition to the above)… the map is calculated based on 2 dimensions, and has no concept of slope (or doesn’t seem to)… excuse my crude illustration. Green line is the actual ground, the map is calculated from overhead based on the exact position (blue). If one large grid is = 1 pass, the software calculates that 6 passes are needed, but the actual distance of the slope is greater as indicated by the green line (also rotated up to the blue line to illustrate the difference in length).

The longer your area / difference in start and end elevation the more pronounced this would be. I don’t have any really big hilly areas, but a good way to test this would be to split that big hill above into multiple pieces and see if it gets better… the largest hill I have is maybe 1/3 of that area and I can essentially get the problem to go away with maxing out the overlap, although not quite.

1 Like

Wonderful news!
Regarding map editing, “split” should have a counter part “merge”.

Where you could select two adjacent areas and merge them into one area.

5 Likes

There has been talk about a virtual merge where you can have two areas treated as one big one by the mower to keep the pattern consistent. I don’t know where that is on the features list. I did a quick search, didn’t spot that one specifically.

2 Likes

I saw a post from co-founder Ken K who said to Scott Grams that it was a great idea and he’d talk to the team. So, I expect it will be a thing sometime.

1 Like

So no obstacle avoidance on back cameras and rain sensor fixes?
I just had a moment where it has been raining 10h and Yarbo still went to mowing. I see sensor value 200+. It should be fixed asap, because it is not usable and can’t trust autonomous mowing.

2 Likes

Rain sensor / detection: Long discussions recently, big one here Rain Detection?

It has been raining here for about three days now as a result of a late season slow moving Nor’easter (some elevations got snow!), and the rain sensor is 75. If I cleaned the sensor thoroughly on a daily basis (not going to happen) and left water on it, it might get above the threshold of 1000, but it dries far sooner than the grass and gets dirty again quickly, and it’s back to registering low numbers even soaked. So I do not use the rain sensor but opt to use the Mowing Operation Initiation Slipper Test, mentioned in one of my responses there, which is an alternative using equipment you may already have in your own home.

1 Like

Appreciate the update. Disappointing that slopes are not a priority, and inconsistent from what your support team said in a response to my ticket.

“ Our team is actively working on it, and it will be optimized in the next firmware update.”

The claims made in Yarbo videos and FAQ are not accurate. Yarbo does not ‘masterfully’ handle slopes. Sure, it can climb and navigate but weaving all over the place and leaving uncut strips is anything but masterful.

4 Likes

They are definitely fixing some of the issues related to the weaving and other issues you describe. Those were software bugs in the latest firmwares.

The dedicated slope pattern for the mower similar to the snow blower will take some more time.

No obstacle avoidance adjustments? Or changes in the wide turns it makes when it bumps into something & has to go around?

I believe all of that is being refined and worked on actively.

This has been my working theory for awhile.

1 Like

I only wish we would get little updates more frequently, put out things that are finished so we could start testing them, a quarter of the mowing season is already over…

2 Likes

Trust me, I think the updates were a bit too frequent before. I appreciate taking the extra time to test it and get it right. It sucks waiting, but it sucks worse when it does something unpredictable and breaks an antenna off or something.

Radical changes definitely need to be slow rolled. Bug fixes I would agree should be pushed out more frequently and out of normal sprint windows.

1 Like

I hear you, but let us, those with the slopes, test it.
Today yarbo didn’t even care about the precise stripes it left out in the past any more, today it seems it was drunk and drove all over the place.


I still ask myself, what where they doing since the kickstarter campaign. Did the problems on uneven grounds never appear in testing. Did they ever test it on the slopes (35°) they were advertising?

2 Likes