Re: [HBRobotics] Safeway Robot Project

64 views
Skip to first unread message
Message has been deleted

Brain Higgins

unread,
Feb 17, 2021, 8:48:30 AM2/17/21
to hbrob...@googlegroups.com
Yes, I’m interested 

Brian

Sent from my iPhone XX

On Feb 16, 2021, at 10:42 PM, Chris Mayer <cmay...@gmail.com> wrote:


Here are videos and photos of the Safeway Robot Project I showed at January's meeting.

Our goal is to build a robot that will go to Safeway, pick up some items, pay, and bring them home as autonomously as we can.

I will set up a meeting shortly after the February HBRC meeting to kick this project off.

If you are is interested please respond PRIVATELY with your best days and times I'll set up our first Zoom meeting.

Please invite anyone of any age and talent you may think would be interested!

Thanks,

Chris Mayer

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/e0f85b3b-ba2e-4b6d-9318-3236228310c0n%40googlegroups.com.
Message has been deleted

Albert Margolis

unread,
Mar 8, 2021, 2:10:41 PM3/8/21
to hbrob...@googlegroups.com
This is a trial. Statistically more likely to fail than succeed. This robot is extremely cute, but I'm doubtful about its ability to navigate many actual streets. Then there are all the social issues regarding safety and how willing people will be to interact with these machines. And I am sure each bot costs WAY more than we would be willing to spend on our projects and possibly more than a grocery store would be able to recover in profits. I am confident that there is still lots of opportunity to be a business success in this segment. But I don't personally care about any of that.

I don't think this trial has anything to do with homebrewed robotics. It is rare that an individual can build a general purpose product better than a well funded team. You build your own for the pleasure of building. Or to get some special feature that serves your purposes just the way you want even though that might make the thing more expensive or generally less desirable. Or because you want to learn some skill involved in the project. 

If you just you want a robot to do some practical things: grab a beer, turn on the TV and wait for somebody else to introduce a product for you to buy. Someday it will arrive at your neighborhood Target. If you want to build a robot, grab some tools and get to work. I think the building process has great personal value, regardless of the result. Building things, particularly robots, is hard. That is what makes it fun.

On Mon, Mar 8, 2021 at 10:36 AM Chris Mayer <cmay...@gmail.com> wrote:
Well, it seems we missed the boat!  Someone already stole my idea.


I will still work on this project, as it was more for fun.  If anyone is still interested in working on this with me, please reply and let me know (no need to reply all), as I would love to not work on this alone.

Chris Mayer

On Tuesday, February 16, 2021 at 10:42:09 PM UTC-8 Chris Mayer wrote:
Here are videos and photos of the Safeway Robot Project I showed at January's meeting.

Our goal is to build a robot that will go to Safeway, pick up some items, pay, and bring them home as autonomously as we can.

I will set up a meeting shortly after the February HBRC meeting to kick this project off.

If you are is interested please respond PRIVATELY with your best days and times I'll set up our first Zoom meeting.

Please invite anyone of any age and talent you may think would be interested!

Thanks,

Chris Mayer

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/5f32a998-844a-4068-8756-01c487005c9fn%40googlegroups.com.

Marco Walther

unread,
Mar 8, 2021, 2:24:48 PM3/8/21
to hbrob...@googlegroups.com
On 3/8/21 10:36 AM, Chris Mayer wrote:
> Well, it seems we missed the boat!  Someone already stole my idea.
>
> https://www.fastcompany.com/90611133/exclusive-safeway-unveils-remote-controlled-food-delivery-carts
> <https://www.fastcompany.com/90611133/exclusive-safeway-unveils-remote-controlled-food-delivery-carts>
>
The article mentions Tracy, CA as one of the first test sites, but I'm a
bit outside the range:-( But I might see them around at some point.

-- Marco

> I will still work on this project, as it was more for fun.  If anyone is
> still interested in working on this with me, please reply and let me
> know (no need to reply all), as I would love to not work on this alone.
>
> Chris Mayer
>
> On Tuesday, February 16, 2021 at 10:42:09 PM UTC-8 Chris Mayer wrote:
>
> Here are videos and photos of the Safeway Robot Project I showed at
> January's meeting.
> https://drive.google.com/drive/folders/13repXQr5AdB3MkcY3iziffBmc4tmwC0F?usp=sharing
> <https://drive.google.com/drive/folders/13repXQr5AdB3MkcY3iziffBmc4tmwC0F?usp=sharing>
>
> Our goal is to build a robot that will go to Safeway, pick up some
> items, pay, and bring them home as autonomously as we can.
>
> I will set up a meeting shortly after the February HBRC meeting to
> kick this project off.
>
> If you are is interested please respond PRIVATELY with your best
> days and times I'll set up our first Zoom meeting.
>
> Please invite anyone of any age and talent you may think would be
> interested!
>
> Thanks,
>
> Chris Mayer
>
> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hbrobotics+...@googlegroups.com
> <mailto:hbrobotics+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/hbrobotics/5f32a998-844a-4068-8756-01c487005c9fn%40googlegroups.com
> <https://groups.google.com/d/msgid/hbrobotics/5f32a998-844a-4068-8756-01c487005c9fn%40googlegroups.com?utm_medium=email&utm_source=footer>.

Screenshot_20210308_112210.png

camp .

unread,
Mar 8, 2021, 2:58:14 PM3/8/21
to hbrob...@googlegroups.com
> ... but I'm doubtful about its ability to navigate many actual streets.
 
    From what I understand these robots from Tortoise are not autonomous but remote-controlled from Mexico City, which is actually pretty incredible. It needs to be training a neural net, however. ;-)
 
Remote operators in Mexico are driving scooters to riders in this Atlanta suburb 
 
Enjoy,
Camp

Albert Margolis

unread,
Mar 8, 2021, 4:04:34 PM3/8/21
to hbrob...@googlegroups.com
I wasn't thinking about autonomous navigation. Look at the diameter of the drive wheels and the wheelbase of the end wheels. I think it will have trouble moving over and around lots of common situations. It might work in new Tracy developments. Not so much in big cities where the density would make them most cost effective or even in lots of older suburban areas like most of the peninsula.
To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/1177365966.497028.1615233402536%40mail.yahoo.com.

Michael Wimble

unread,
Mar 8, 2021, 6:03:36 PM3/8/21
to hbrob...@googlegroups.com
I’m wondering what they think is going to happen with people on sidewalks. There is little enough space on sidewalks in cities even now. I suspect people are not going to like making way for these things in a lot of situations 

On Mar 8, 2021, at 1:04 PM, Albert Margolis <almar...@gmail.com> wrote:



Brian Higgins

unread,
Mar 8, 2021, 7:06:40 PM3/8/21
to hbrob...@googlegroups.com
Mather VA in Davis has had a med cart robot for years.  Everyone got along.
One of the first VA hospitals.

Brian

Sent from my iPhone XX

On Mar 8, 2021, at 3:03 PM, Michael Wimble <mwi...@gmail.com> wrote:



Chris Albertson

unread,
Mar 8, 2021, 7:07:23 PM3/8/21
to hbrob...@googlegroups.com
I think the idea to work on a robot that can go to a store and buy an ioitem then return it is a GREAT goal.    Even if no one wants such a robot or if some one else already built it.   The point here is (I assume) to set the bar high and learn about robotics.

Some suggestions,
1) Assume the world is near "ADA Compliant". This means streets and stores are set up for access by wheelchairs.  This assumption will fail in many actual locations but in US cities there are MANY places that are ADA Compliant.   Enough places that the robot can be tested.  (Another use for this robot might be automated ADA Compliance testing.)

2) the project is decomposable to at least (a) grasping and lifting, using vision and other sensors, (b) payment, (c) navigation.   Someone could work on one part without needing a full-up robot.   There are more parts.  Some interface documents would be needed.

3) what we need to be looking at is more commonly thought of as "self-driving car technology" than as a "robot".   Even if the terms are technically the same.  Delivery is exactly like driving only much slower so there is less risk and less computational power needed.  It is easier to drive at 1 MPH than at 65 MPH but the technique is the same.     We might want to look at Autoware.  Autoware is built on top of ROS2 but is specialized for cars and outdoor driving.  It's reasonably mature.

See here for details.  https://www.autoware.auto/

4) What's needed to kick this off is a better communications system the combines messages, CAD and version control, and maybe a wiki.   No platform has this so it needs to be a combination.   If I were picking, I'd say use Github for version control, wiki and issue tracking and Onshape for mechanical design as Onshape has very excellent data handing and keeps its files in the cloud.

5) I have a chassis I think would work well and is easy to buy.  I have one.  It is easy to strip this little car of excess decorations and replace the steering wheel with a stepper motor and reduction drive.   The payload is over 100 lbs so there is room for a much larger battery, sensors, and a few computers.     There are available cheap on Bay and you can buy parts.
https://www.walmart.ca/en/ip/power-wheels-disney-frozen-wrangler-jeep/6000188920637

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/CAPFBGAd2A32ZYPCZs-1nfJ2wQoSvao9FzLbN0_OwhOiiwwKtjw%40mail.gmail.com.


--

Chris Albertson
Redondo Beach, California

Rafael Skodlar

unread,
Mar 9, 2021, 12:16:45 AM3/9/21
to hbrob...@googlegroups.com
Very good Al!
I was also wondering about the social and health issues with robots running to the grocery store and back. Going to the store is a small social event for many, me included. It's a chance to meet people on the same street, learn about their cultures, get local news, and events. Why replace it with a machine full of components that were built around the world where pollution is not a concern to manufacturers? What about electricity, bad weather, snow, wind, floods?
The worst thing at this point are large monopolies that are taking over most business activities everywhere. They are destroying small businesses that used to be run by people who love their work. Old book stores, electronics stores, etc. are disappearing. Frys is gone.

Local home improvement store Lowes used to have a robot moving between the isles. Last time I remember seeing it, it was standing idle with an LCD display showing Ubuntu desktop. That robot is gone.

I can envision robots on the sidewalks only when helping old or disabled people go around or cleaning the streets. No need to deliver stuff to lazy dopes!
It seems to me that we need a better definition for practical use of robots. Let's design robots for dangerous jobs in mining, demolition, construction, handling biohazard and chemicals, farming, etc. 

For inspiration or ideas related to robots I recommend the following video from Japan:

If that's not enough check this:
"San Francisco Spends $30 Million Cleaning Feces, Drug Needles"
What do you prefer robots to do, cleaning dirty streets or have them deliver groceries, needles, and poop collected on their wheels to your front door?
From golden state to dirty state in one punk generation.

Rafael


To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/CAPFBGAedabbNv-mBp3OSKu6%2BLWwzjkncr416eHbwpjpCRsbg3g%40mail.gmail.com.


--
Rafael

Brain Higgins

unread,
Mar 9, 2021, 9:36:58 AM3/9/21
to hbrob...@googlegroups.com
As a visually impaired person with a white smart cane I can tell you about sidewalks!  Hey sidewalk is an excuse to put telephone poles, signs, park benches, trash cans, bus stop.

Brian

Sent from my iPhone XX

On Mar 8, 2021, at 11:58 AM, camp . <ca...@camppeavy.com> wrote:



Oran C

unread,
Mar 9, 2021, 12:48:27 PM3/9/21
to hbrob...@googlegroups.com
I actually live near Windsor, Sonoma, ca. So I'm excited to see these bad boys in action. 
Also yeah they are probably really expensive. 



To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/FE9F690E-140B-479B-A8BD-0C3D7D37C6DE%40comcast.net.

Chris Albertson

unread,
Mar 9, 2021, 2:12:29 PM3/9/21
to hbrob...@googlegroups.com
Today many people are having groceries delivered to their porch.  My son had been doing this before COVID-19. The pandemic caused a big uptick in business. Many people think the food delivery business will be with us now for good.  But it was with us before.

The other thing the pandemic has taught us is the many people can actually work productively from home

Put these two things together and you have an economic justification for food delivery that is sustainable.   What if you can make $50 per hour at your job, working at home.   (this is the low end, many people make a lot more)   Then what if someone will shop and drive to the store for you for $20 per hour.   You pocket $30 for every hour someone shops for you.

Yes, of course, many people can't make $50/hr in any job and can't work at home. Some people are retired and some are students.  But we have a huge segment of the population now who can pocket $30 per hour by using a shopping service.

There is another robot delivery service in Japan, outside of Yokahama.   I've been some years ago. It is an upscale suburban area. I've been there myself and can see why they used this place for testing.  People there can afford to pay for delivery and the streets and sidewalks are not like in central Tokyo.   I think there are enough places where this could work that it makes sense and then in other places a different system is used.   They are doing the "low hanging fruit" model of development.

Here is an article I found for a different test area.    The Japanese post office is also using the same robot from Panasonic.    But these are primitive robots, mostly teleoperated.  The purpose is just testing https://www3.nhk.or.jp/nhkworld/en/news/backstories/1443/

The purpose of the test is to see if the robot can drive on existing streets and sidewalks and if customers like the service.   Panasonic has deep pockets and can afford to do this at a loss for a long time, gambling that it will pay off in 10 years.



To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/CAB4wyt01Vz33AvQBvaEp8%2B1Zx9EM-OX1mNxhmmEFaTR1-XoRww%40mail.gmail.com.

Chris Albertson

unread,
Mar 9, 2021, 2:12:35 PM3/9/21
to hbrob...@googlegroups.com
Brian,

Yes, sidewalks have all those obstacles.   I keep thinking what is needed is an electronic guide dog that talks.   The machine could see and identify what is there and unlike a real dog could read printed signs and Google Maps to know how to get to any location in the world.

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/FE9F690E-140B-479B-A8BD-0C3D7D37C6DE%40comcast.net.
Reply all
Reply to author
Forward
0 new messages