[HN Gopher] Launch HN: Charge Robotics (YC S21) - Robots that bu...
___________________________________________________________________
Launch HN: Charge Robotics (YC S21) - Robots that build solar farms
Hi HN! We are Banks Hunter and Max Justicz, founders of Charge
Robotics. We make robots to automate the most labor-intensive parts
of solar construction. We just got back from our first demo on a
150MW solar construction site in Iowa, where we showed off our
initial prototype: an autonomous forklift that unloads pallets of
solar modules from a truck and stages them around the site. It's a
huge milestone for us, and we felt like now would be a good time to
share what we're working on more publicly. You can see a couple
videos of our robot in action here: Staging modules on the site in
Iowa: https://youtu.be/Fwf4v8upuoI Performing a two-pallet sliding
and unloading operation in our warehouse:
https://youtu.be/EOJiyMXpVeQ As solar modules have become
commoditized and prices have plummeted, solar has become the
cheapest form of power generation in many regions. Demand has
skyrocketed, and now the primary barrier to getting it installed is
labor logistics and bandwidth. Every solar construction company
we've talked to is drowning in demand and turning down projects
because they don't have the capacity to build them. 1/5th of all
the solar that exists in the US was installed last year! We're
engineers who have been friends since living together at MIT where
we studied robotics and CS. We always wanted to start a company
together. We zeroed in on solar after seeing compelling statistics
about its cost effectiveness and projected growth - and because we
shared a motivation to do something about climate change. We
actually started out writing software to predict optimal locations
for solar sites (searching land for sale and scoring by price,
amount of sunlight, proximity to existing substations) when we
decided to learn more about what comes next. Utility-scale solar
farms (2MW+) are mechanically quite simple. They feature a steel
racking system held to the ground by vertical posts ("piles"), and
overwhelmingly (90%+) feature a single motorized axis to track the
sun over the course of the day. Modules are then fastened to this
axis with brackets. We're using a two-part robotic system to build
this racking structure. First, a portable robotic factory placed
on-site assembles sections of racking hardware and solar modules.
This factory fits inside a shipping container. Robotic arms pick up
solar modules from a stack and fasten them to a long metal tube
(the "torque tube"). Second, autonomous delivery vehicles
distribute these assembled sections into the field and fasten them
in place onto target destination piles. This is a hard technical
problem, but not research-level hard. We think of it as the
"homework version" of self-driving cars, as we're operating in a
semi-structured environment (flattened dirt field) with drastically
fewer edge cases. Manual construction today breaks about 0.1%-0.5%
of modules during installation, which is an easier bar for us to
target than the stringent performance requirements of the AV world.
We're operating in a risk-averse industry, though, which makes
deploying new technology more challenging. One industry-standard
term we've become very familiar with is "bankability". It's
difficult for projects to secure funding from lenders if they
aren't using parts that have already spent years out in the field.
We've seen surprisingly little penetration of technology into this
space in general. Projects are largely tracked with sticky notes in
a "command room", material delivery schedules are highly volatile
and often not known until days in advance, and there's no live
monitoring of construction progress, making current status opaque.
We actually had a site we visited outright lose a forklift - we
were surprised that all vehicles aren't GPS tagged and monitored,
especially given they're operating on multi-thousand acre sites.
Our system is the first to handle the full mechanical installation
of existing solar components (remember bankability). We've tweaked
the order of construction operations slightly to be more robot-
friendly, as the more precise operations involved in fastening
modules to steel tubes happen in a more controlled factory
environment. For our mobile robots, we're building on top of
existing vehicles (called telehandlers, or reach lifts) which are
already ubiquitous on these sites due to their enormous tires and
broad capabilities. They're able to unload shipping containers due
to their extending boom, as well as move materials around the site.
On our prototype vehicle, we did some significant up-front reverse
engineering including mapping out CAN messages sans documentation.
The steering and brake were directly hydraulically actuated (no
drive-by-wire), so we added motors to both in order to control them
with our software stack. The most unique sensor we contributed was
an optical mouse sensor mounted onto the boom joint, telling us the
extension distance. The backbone of our robotic sensing is a
robust vision system. We're using stereo cameras for SLAM and
object detection. Fortunately, solar construction sites already
have detailed engineering drawings including GPS coordinates of
each vertical post in the ground, so we have a detailed map of the
site to localize ourselves on. Watching the existing process for
large-scale solar installation in real time evokes the sense of
watching paint dry or grass grow, only it involves hundreds of
workers. After witnessing the physically grueling and inefficient
process of workers manually installing thousands of solar modules,
we realized there had to be a better way of building solar, and
that increased automation was the way forward. Our goal is to
transition the world to renewable energy as quickly as possible.
We're excited to share what we're working on with HN - please let
us know what you think in the comments and we'll be around to
respond! P.S. We're hiring! If you want to work on cool robots
with a positive climate impact, please reach out:
https://chargerobotics.com/careers.html
Author : justicz
Score : 83 points
Date : 2022-03-23 16:47 UTC (6 hours ago)
| gridspy wrote:
| Amazing. Almost makes me want to move to the states (from NZ) to
| work with you. Always gratifying to see a job posting where you
| tick every box.
|
| What a great automation environment! Almost clean-slate too, you
| don't need to integrate with a patchwork of other solutions that
| are different per site.
|
| Robots as a service is definitely the way to go. Much lower
| apparent risk for your clients, less support costs for out of
| date product you sold early on... many great reasons.
| veb wrote:
| Mate, why don't you start something like this in NZ? We totally
| need more robotic startups IMO.
|
| Years ago I 'worked' at Scott Technologies, around the time
| they were making this 'smart' straw making machine that had the
| chocolate/strawberry flavour inside the straw and you just put
| it in some milk and bam. Really cool going from planning to
| actually testing the robot (was for some client, can't remember
| who). I did this while on a high school placement programme in
| year 13.
|
| I really love robotics, but I moved in a different direction
| after that!
| ParrotAI_Inc wrote:
| This is awesome! That Iowa video is very cool.
| somethoughts wrote:
| Cool stuff! Would love to see a drone shot and/or time-lapse of
| an solar farm installation.
| justicz wrote:
| We have a video with some drone shots of pallet
| unloading/staging here!
| https://www.youtube.com/watch?v=d9MLKmfmR30
| manholio wrote:
| This seems to be the same age-old problem of construction
| automation. A system that can really deliver for solar farms
| could be adapted for many other construction types, at least in
| principle. Since the world has largely failed to automate
| construction, and not for lack of trying, you really have to
| wonder if a startup could really disrupt the field, and I say
| that as an automation engineer.
|
| The Construction Physics blog is mandatory lecture on this
| subject. You will find the many and varied ways huge companies
| failed at automating construction, and why construction is just
| different than other fields:
|
| https://constructionphysics.substack.com/p/where-are-the-rob...
|
| https://constructionphysics.substack.com/p/why-did-agricultu...
|
| https://constructionphysics.substack.com/p/japans-skyscraper...
| cmuell89 wrote:
| I disagree that this is the same age-old problem. Automation in
| construction has failed more often than not in scenarios that
| require a high-level of engineering and/or exhibit complex
| tasking on-site. Your second listed article provides a good
| rationale for the failure of automation to infiltrate
| construction: mechanization. Many of the on-site build
| requirements for construction application are much more
| complicated and intricate than some of the motions required for
| agricultural automation. Furthermore, process design in
| construction is human-centric with a wide gap to transition to
| automation-centric design. This is well-outlined in your third
| article e.g. setup/tear-down costs, robot unfriendly
| environments, start-stop processes.
|
| However, Charge Robotics has a problem that is generally free
| of much of the above hindrances and is employing a modular
| approach. Modular construction can lend itself very well to
| automation. See the modular housing boom that employs
| automation quite heavily.
|
| One of the more complex construction portions of the build is
| handled by a portable factory:
|
| "First, a portable robotic factory placed on-site assembles
| sections of racking hardware and solar modules. This factory
| fits inside a shipping container. Robotic arms pick up solar
| modules from a stack and fasten them to a long metal tube (the
| "torque tube")."
|
| This process could also employ human labor as well. The rest of
| the build process requires a more simple transfer of these pre-
| built components with a final assembly:
|
| "Second, autonomous delivery vehicles distribute these
| assembled sections into the field and fasten them in place onto
| target destination piles."
|
| The locations of operation are large open fields with minimal
| occlusion (which often makes robots in many construction
| settings infeasible) and reuses prior equipment (bankability
| and minimizes setup/tear down costs).
|
| I don't think this is an easy problem as there are significant
| technical engineering requirements, but if each modular process
| can be proven effective, then the external factors that
| prohibit automation adoption in construction outlined above are
| fairly minimal.
| justicz wrote:
| Thanks for putting this together! This is basically what we
| would've written. There are many aspects of solar
| construction that make it a much more robot-friendly
| environment to operate in than that of a generic construction
| site (detailed engineering drawings with GPS locations of
| piles, graded terrain, low number of obstacles). And as you
| mentioned, we're eliminating a lot of variables by bringing a
| portable factory onto the site.
| DreamFlasher wrote:
| How can I invest?
| mrraj wrote:
| Interesting idea - having worked in the back office of a solar
| construction company for a number of years I can see a few
| challenges.
|
| Install labor is relatively cheap but more importantly flexible -
| relative to the cost of the project install is low double digits
| %. If a job isn't running smoothly you cut your temps and move
| your main crew over to another site or send them home if they're
| local. I would wonder how this work fit in with an automated
| build/install solution.
|
| Unless you have control at the GC level you're going to be
| start/stop with electricians cutting trenches in front of you,
| missing materials (tariffs/port strikes), permitting delays, etc,
| and you'll have expensive idling equipment that's tough to move.
|
| I get that this is the problem you're trying to solve - but I'd
| definitely suggest going to enough bread and butter 2-10MW sites
| where this sort of thing is more common. Also keep in mind
| Central CA in summer is not Massachusetts in November, weather
| makes all this 10x worse.
| banks_h wrote:
| Definitely, we've experienced workers being sent home for a
| number of reasons (too cold, too windy, too rainy, materials
| didn't show up as scheduled). Fortunately with an autonomous
| system there's no logistical overhead, if some external factor
| prevents us from working then we aren't stuck covering hundreds
| of per diems and sending everyone home. Yes, we'll have some
| equipment depreciation as it's not working, but it's at a
| similar rate to the rented or depreciating trucks/skid
| steers/telehandlers on existing sites when no one's working
| them.
|
| Totally hear you on the CA vs MA weather swings (I've lived in
| both!). We actually see this as a key advantage for us. Sites
| we've visited in central CA have significant water distribution
| logistics, frequent shade breaks, and still have high heat
| exhaustion rates. Sites we've visited in northern climates
| report about 50% speed reduction in work due to gloves/cold,
| and will send workers home if the temperatures drop far enough.
| Our robots can handle both climates without an impact on
| installation rates (or any of the associated health risks for
| workers!).
| hutzlibu wrote:
| "Also keep in mind Central CA in summer is not Massachusetts in
| November, weather makes all this 10x worse"
|
| Important to keep in mind, but I would suspect(and hope), most
| solar farms will be installed, where it is sunny and dry.
| mrraj wrote:
| Mostly this is correct. Few exceptions -
|
| Weather patterns on East coast can be unpredictable, much
| more so than central valley CA. torrential rain = mud leading
| to bogged down equipment, this can be the difference between
| a profitable job and unmitigated disaster for an installer.
|
| Second, construction cycles in utility solar can be a bit
| wonky because of ITC tax credits, lots of turnkey providers
| looking to have construction starts in q4.
| bckr wrote:
| Hey guys! I'm so excited to see your launch HN! We chatted about
| your who's hiring thread middle of last year when you were in
| stealth mode. I still think this is such an epic idea, very happy
| to see you succeeding.
|
| Can't wait to see what comes next!
| banks_h wrote:
| Thanks! Great to hear from you (assuming I don't miss my guess
| from your username), hope everything's been going well for you.
| Looking forward to staying in touch as we make progress!
| atlasunshrugged wrote:
| This is super cool! I'm curious how much of a problem the policy
| side of things is here? My impression is that some of these
| projects fail because of NIMBYism and long lawsuits over land use
| and the like (despite the net positive climate impact).
| justicz wrote:
| This definitely comes up. Some communities are more receptive
| to solar than others, and it is important that the local
| community buys into a project before it gets built. We've
| definitely passed by signs in the lawns of local residents
| unhappy about the project being built next door.
|
| We are fortunate in that those sorts of problems are actually
| largely the responsibility of the solar project developer,
| whose job it is to finance the project, get land rights, do the
| appropriate environmental/interconnect studies, and build
| relationships with the community. The developer lets the
| construction company (and us) focus on building the project.
| But it's an excellent question.
| cinntaile wrote:
| Since you say it's a risk averse sector. Do you think you will
| need to do the entire process (building the solar farm) or do you
| expect that companies that build solar farms will hire your
| robots for part of the process? Or are you aiming for an
| acquisition?
| banks_h wrote:
| We're operating as a subcontractor for existing large
| construction firms, where we come in and do the mechanical
| construction step.
|
| In general robotics companies need to pick between "sell
| robots" or "operate robots as a service", and we've picked the
| latter model. Personally I think that throwing robots over the
| fence to a customer typically results in more headaches on both
| sides. Instead, we deliver mechanically assembled solar farms
| to construction companies, and they don't need to worry about
| how we got there - just that we did it quickly, efficiently,
| and spared them the staffing logistics.
|
| Re:acquisition, we believe we can become a significantly
| profitable company operating independently. If we can become
| the best way of building solar, we'll be in a great position to
| sell services to all of the existing large solar construction
| companies.
| avip wrote:
| Love that Bill Withers soundtrack!
|
| I also work in the "robots for PV" space, it's just the place to
| be. Good luck!
| transistor-man wrote:
| Congratulations Banks and Max, this is excellent!
|
| -Dane
| banks_h wrote:
| Dane! Good to hear from you, and thanks!
| sklargh wrote:
| Solar Factorio! Love it.
| justicz wrote:
| :) Fortunately there are way fewer biters.
| joefigura wrote:
| Awesome to see your progress! Good luck deploying more solar!
| monkeydust wrote:
| Automation + Solar two of my favourite things, honestly! Congrats
| on getting this far.
|
| Have you considered, or have views on, the scope for using
| telerobotics in this field as a step to full automation? Point
| being the qualified labour could be anywhere in the world for a
| particular job and the data generated from that could be used
| towards the drive for full automation.
| justicz wrote:
| Teleoperation is on our radar, especially for getting robots
| un-stuck when they are confused about a particular situation.
| Our vision is that a single remote operator can monitor an
| entire fleet of robots, stepping in to resolve confusing
| situations as necessary. We've seen this model deployed at
| other robotics companies like Third Wave Automation [1] and
| Teleo [2].
|
| [1] https://www.thirdwave.ai/ [2] http://www.teleo.ai/
| ewjt wrote:
| Aside from labor _replacement_ and the implied capital
| efficiency, are there other benefits automation can bring to
| solar installations?
| banks_h wrote:
| Absolutely. A couple that we're closely looking at:
|
| Better documentation/data/insight into exactly what was built
| when, install rates, fuel consumption, and vehicle utilization
| during construction. Having our robots take a timestamped
| picture of every bolt that's installed is a significant win for
| construction companies in case any warranty issues come up
| after install. Currently there's not much visibility into
| exactly what was built where/when, so concrete data here will
| help significantly with scheduling/planning. We can actually
| generate a live updating map of the solar farm as it's being
| built.
|
| Nighttime operation. Currently, all tasks on the site are
| performed during the day, so any amount of work we can shift to
| night time reduces site congestion. This actually can speed up
| the overall installation, as right now there's limited space
| between rows of panels, and multiple teams/vehicles trying to
| navigate through them simultaneously. We think of nighttime
| operation as analogous to parallelization as tasks that workers
| were waiting on can be performed asynchronously.
| marcell wrote:
| Is labor cost a significant obstacle to building solar farms?
| What % of initial capital goes to labor?
| rw2 wrote:
| This is the real question
| iancmceachern wrote:
| From my experience the biggest cost other than Panels and
| copper wire is the framing and structural infrastructure.
| Building codes, wind loads, etc. make it so its not as simple
| as propping up a few panels and plugging them in. It's
| significant and also cost inefficient in many ways.
| mdorazio wrote:
| Not really from what I've seen. There's a rabbit hole of
| research reports, but the key chart is shown here:
| https://www.solarreviews.com/blog/what-is-a-solar-farm-do-i-...
|
| Maybe 10-15% of total cost depending on install type.
| banks_h wrote:
| Reports are somewhat varied here but we find that about 10-20%
| of the overall site cost goes into labor. The bulk of the rest
| is component cost for solar modules and tracking systems, which
| are relatively commoditized and so are a thin margin above the
| cost of the raw materials to make them (silicon, steel). Those
| costs won't drop unless the underlying materials get cheaper.
|
| We find that beyond the direct cost, the logistics of getting
| hundreds of people to a remote site is a significant challenge
| for construction companies. We directly saw how much of a
| hurdle this can be when we were on a site in Iowa with the
| closest city an hour and a half away. Weather conditions are
| also a factor here, work on the site was actually shut down due
| to conditions that were too cold for laborers.
|
| As we've interviewed solar construction companies, all of them
| have told us that they're turning down potential jobs just
| because they don't have enough people/capacity to build them.
| Unfortunately they can't simply raise their hourly rate by a
| few dollars to make this problem go away, because the
| challenges are more the regional specificity and logistical
| challenges of getting workers to sites.
| toomuchtodo wrote:
| Is the solution than perhaps the rapid build up and turn down
| of mini cities for workers during the build phase? The
| problem sounds like Black Rock City public works for solar
| projects, versus robots for labor needs that can change
| rapidly (trenching, racking install, panel mounting, etc).
|
| TLDR Worker housing and logistics vs robotics.
| banks_h wrote:
| Black Rock City's DPW for solar is a really interesting
| idea! Temporary housing is definitely used at times for
| more remote projects, but it's more expensive than just
| booking out all the local hotels (of course, once you've
| saturated hotel rooms you might not have the option).
|
| Fortunately the labor needs are typically predictable, with
| each step you mentioned happening in parallel across a site
| spanning 100's or 1000's of acres. One of the main reasons
| that larger sites are particularly appealing for us is that
| the work done by our robots is continuously in demand
| somewhere, as opposed to a small site which might move in
| more discrete stages and have the rapidly changing needs
| you mentioned.
| dmead wrote:
| Should we do this? My understanding is that solar installation is
| the next (semi?) skilled labor that could potentially be careers
| for many people.
|
| It doesn't seem to be the most socially stable thing to cut out
| humans from this job market before it really booms.
| justicz wrote:
| It's a good question. When you look at why utility-scale solar
| construction isn't happening fast enough, you quickly find that
| there are fundamental scaling problems associated with getting
| hundreds of people to very remote construction sites every day
| for the duration of a project -- and these problems get worse
| as the sites get larger and more remote. While our robots will
| reduce the number of people required to build a given MW of
| solar capacity, increasing the labor supply will dramatically
| grow the number of solar installations happening overall each
| year, and with it create a lot of skilled construction jobs.
|
| > My understanding is that solar installation is the next
| (semi?) skilled labor that could potentially be careers for
| many people.
|
| This is partially true -- a lot of the more specialized solar
| jobs show up in the residential solar market, which is growing
| quite quickly (residential solar grew 30% in 2021 over 2020).
| The training period for the tasks we're performing on large-
| scale sites is a couple of days.
___________________________________________________________________
(page generated 2022-03-23 23:00 UTC)