(Unedited) Podcast Transcript 406: Sidewalk Delivery Robots

October 26, 2022

This week we’re joined by Amanda Howell of Urbanism Next at The University of Oregon to talk about Sidewalk Robots. We chat about their study on sidewalk delivery robots with the Knight Autonomous Vehicle Initiative and what they learned about community engagement and small delivery vehicles on city sidewalks.

For the podcast audio, visit Streetsblog USA or our hosting site at Libsyn.

Below is a full unedited transcript of this week’s podcast.

Jeff Wood (1m 12s):
Well, Amanda Howell, welcome to the Talking Headways podcast.

Amanda Howell (1m 48s):
Thank you so much for having me. I’m excited to be here.

Jeff Wood (1m 51s):
Well, thanks for being here. Before we get started, can you tell us a little bit about yourself?

Amanda Howell (1m 54s):
Yeah, so I am a researcher with the Urbanus of Next Center at the University of Oregon. Although I’m based in Portland, I have been doing this for almost five years. I do research on topics related to broadly new mobility and their impacts on cities. And so I’ve done research on micro mobility equity related to micro mobility, autonomous vehicles, TNCs or transportation network companies like Uber and Lyft, kind of everything that sort of falls under this new mobility umbrella and making sense of it. And really not from the standpoint of like what is the technology itself, but more what are the impacts, what are the benefits, what are the challenges? Is this something that we should be incorporating into our cities and is it actually helping people?

Amanda Howell (2m 36s):
That’s the work that I do.

Jeff Wood (2m 37s):
I like that because when I read a lot of articles, especially when they’re from kind of the trade papers of the technology industry, they’re just like, This is what this company does and this is how much money they raised and this is, you know, what the future is gonna be like according to their CEO who’s very optimistic and is trying to raise money. And it’s like, you know, I wanna know what the impacts are. I wanna know what the situations are where this is actually useful versus you know, your kind of rosy outlook about your company.

Amanda Howell (3m 2s):
Yeah, exactly. What are the use cases? Who is benefiting, you know, who, who maybe isn’t set up to benefit? If it’s something where there is the opportunity for a lot of people to benefit if certain changes were made, are these changes that we should make? You know, to what extent should we be helping to make sure that these services can be deployed?

Jeff Wood (3m 21s):
Yeah. Well, so how did you, what was the first kind of introduction to micro mobility or the technology infiltration of transportation in general? What was like your first introduction to the topic and and how did you get into it?

Amanda Howell (3m 32s):
Well, so I grew up in Los Angeles, so you know, in the very car dominated culture at the time, I didn’t really realize that life could be different. I didn’t realize that I didn’t have to entirely rely on getting a ride for my parents, you know, until I could drive myself around that, that was even an option. And so then I moved to the Bay. I was in the bay for a long time. I went there for my undergraduate education and started getting, you know, more familiar. I lived there without a car. I started walking more. I started, you know, riding a bike around San Francisco and taking transit way more than I ever had in la. And then what really actually kind of moved me into the transportation sphere. So I, I went back to school to study urban planning.

Amanda Howell (4m 12s):
And before I went back to school, I had been working with an organization in the Bay Area called, it’s called Mount College now. It was the Prison University project when I was there. And I worked with people who were incarcerated, but many of them were students were who were getting out. And the incredible challenges that they were facing in terms of, you know, not only finding jobs and finding housing, but getting to the places where, you know, where they were able to find jobs and where the housing was located. You know, it was so incredibly challenging for them to be able to navigate and move around. And I think that that for me was sort of this like kind of clicking moment of like how, you know, I had been really interested in housing and all these other issues and just like this idea that like, oh my God, like it seems so obvious, but like how do we get places?

Amanda Howell (4m 59s):
How do we, how do we get to the places we need to go and how easy is it and why have we made it actually in so many places in new life? So incredibly hard. That is where my interest in this all kind of stems from. Now I live in Portland. I thankfully am now on a place where like I barely ever drive. I bike commute, I occasionally ride scooters, I ride transit, you know, and just being able to live in a place that has that level of access, I think that just has made me an evangelist for, you know, for the fact that I enjoy my life so much more when I spend less of it in the car. Although I recognize, I also want to be mindful of the fact that we, you know, we have created cities and we’ve created the way our urban form is that, you know, for many people a car is I think in some ways a stop gap measure until we start to do a better job of creating places that are better connected with better, you know, more convenient fast transit.

Amanda Howell (5m 54s):
You know, a car probably does have its place, but I’d like to invest to continue to move towards, you know, a future where we, people don’t have to rely on a car to get to where they need to go. And that micro mobility can be a component of it. Cause I think that, you know, I, I had fun, I love riding.

Jeff Wood (6m 13s):
I’m interested in the idea of incarceration for a long period of time, and then getting out and trying to get places and trying to understand systems that might have changed a lot by the time you get out. And especially from a technology standpoint. If you think about, back to 2009, I guess when iPhone came out, that’s only so many years and there’s been a huge amount of changes. And then there’s been people I, I imagine that have been in prison since that time. Oh yeah. That when they get out, it’s like, what in the heck happened? Like I went in and the Backstreet Boys were doing their thing and then I’m out and there’s a crazy amount of technology change.

Amanda Howell (6m 44s):
Absolutely. Yeah. I mean, I was working with folks, you know, this was twenty thirteen, twenty fourteen, you know, iPhone had only been out like seven years. And I was working with people who’d been incarcerated sometimes upwards of 20 years. So, and you’re living, I mean I think that being in that space, you’re living in this microcosm you, you’re basically moving around between buildings, everything’s, I mean, pretty much everything’s on foot from one place to another. And for a lot of those folks, the only times when they’re even getting into a vehicle is, you know, if they ever been taken out for a medical, going out to a hospital or something like that, or being moved between facilities. So to then all of a sudden be, you know, sort of given gate money and then told to go on your way and have to navigate this system that is incredibly complex with new technology to adopt.

Amanda Howell (7m 30s):
And, you know, a lot of challenges in terms of like getting all the things that you need in order to operate in that world where you might not have your ID driving is not an option anyways. But then you also don’t have, you know, a debit card. So like the issues with cash and, and not being able to get on the bus and all this stuff. I mean, it’s so complicated.

Jeff Wood (7m 48s):
Yeah, sounds very difficult. Let’s talk about delivery robot, shall we? Yeah,

Amanda Howell (7m 53s):
Let’s pivot to that.

Jeff Wood (7m 55s):
Well, so this project was initially supposed to be with autonomous delivery vehicles, but ultimately because of the pandemic, you decided to go with Sidewalk Robots. I’m, I’m wondering how much thinking changed during the pandemic because delivery became such a big deal

Amanda Howell (8m 8s):
So that it was a huge component of it. So originally the projects, this is for the night autonomous vehicle initiative that has been underway since 2018. There’s four cities involved. It’s Detroit, Pittsburgh, Miami Dade, and San Jose. And originally these four cities were selected because they had a lot of autonomous testing and deployment and they also had pilots that were being planned to test autonomous vehicles. And actually at the time when everything was starting, the idea was that they were primarily gonna be testing autonomous passenger vehicles. And for a variety of reasons, the pandemic is one huge component of it.

Amanda Howell (8m 48s):
But I think that’s actually just one component. The, the pilots started getting shelved or permanently shelved as part of the function of the pandemic. But also I think the business environment, you know, the technology, there’s been lots of articles I think written where it’s like in 20 17, 20 18, it was like, autonomous vehicles are gonna be everywhere tomorrow. And then as we now know, of course that’s not true and that it is a lot more complicated. And so it was kind of both of those things happening together in tandem that ended up, you know, we then in 20, early 2020, the pandemic’s underway, the pilots that we’ve been planned had been working towards, they were being shelved. And so we had these conversations around, you know, okay, like people are at home.

Amanda Howell (9m 30s):
We’ve seen this huge uptick in deliveries. Is this a technology that we can test with personal delivery devices or sidewalk robots and see what does this look like in terms of actually trying to deploy these on city streets? Because for the most part, the delivery robots up until that point had largely been deployed on college campuses. You know, that’s where they’d been. But you know, at the same time, I think another pressure that was underway in 2020 and a little bit in the years before that was that a lot of states had started passing bills legalizing the operations of these devices on city streets. And I think there was kind of an increasing sort of concern from city perspective that, you know, we don’t really don’t know that much about how these devices operate and yet they’re being legalized at the state level saying they can operate.

Amanda Howell (10m 16s):
And so we really, we really feel like we need to learn more about, you know, what is that, what do they actually look like? Can they navigate our streets and infrastructure as they currently exist? Who stands to benefit? Who’s maybe gonna be left out if we deploy these? And so there was just a, there was like a lot of driving forces that came into this idea that like, okay, this seems like a good time for us to try and pilot these delivery devices. And then, you know, reaching out to different providers of the technology and kind of figuring out who would be willing to do this, experimenting with us. Huge props to Kiwi because I think this was a big question mark of like, okay, we want to try deploying these devices on city streets. And not only that, we also really wanna work together and we wanna try deploying them and testing out use cases that you as a company might not otherwise do.

Amanda Howell (11m 3s):
You know, we wanna try testing delivery farm boxes to low income residents in San Jose. That’s probably not a use case they might have done on their own, right. So it was just a big experiment was really what it was that we did for six months in these four different cities.

Jeff Wood (11m 17s):
Well, so why did the Knight Foundation want to get involved in this? Obviously they have a certain view of what research they wanna do, what initiatives they wanna get behind. I’m curious what their interest was in this.

Amanda Howell (11m 27s):
So the Knight Foundation has a couple of different big topics that they focus on. You know, obviously journalism is a huge component of it because this is the Knight family. The Knight Brothers had all the newspapers and three of these four cities. So the only one that isn’t is Pittsburgh. They are all night cities. So they’re cities where the night family had newspapers. And so, you know, journalism is a big component of it. They also do a lot of work around public space. But one of the other things that, another kind of pillar of work that they’ve been starting to focus on more and more is sort of smart cities and the idea of like, how are we engaging community around technological change? That’s been an, an area of interest for ’em.

Amanda Howell (12m 9s):
And I think, again, kind of going back to like, I guess this would’ve been sort of late 2017, early 2018, the, you know, they had three cities within the night sphere, San Jose, Detroit, Miami, were all sort of in this place where there were a lot of AV companies that are operating there, they’re testing there, they’re creating, you know, test facilities. And all of these cities were also starting to play around with the idea of doing these AV deployments. And I think the Knight Foundation was sort of, you know, from their perspective was feeling like, okay, this is a really interesting place that’s starting to develop, but we really feel like what’s missing from these conversations so far is like, where’s the community engagement?

Amanda Howell (12m 52s):
What’s the community voice in this conversation? And so that was, that was what the driving force behind the idea that they were going to invest in this initiative. And so, and then Pittsburgh, they added, you know, even though it’s not in the night city, because Pittsburgh, I mean there’s so much testing with Carnegie Mellon, the deployment of, and creation of the technology that it felt important to include them as part of the cohort. So that was the driving factor. And so the funding, the grant funding that they gave is really meant to fund community engagement around these topics and around autonomous vehicle technology.

Jeff Wood (13m 24s):
You did do a lot of community outreach for this. I’m wondering what the results were in terms of reaching out to people. How did you reach out to people first off, and then also like, you know, what kind of results are, what kind of feedback did you get when you told them about these little robots that were gonna go on the street?

Amanda Howell (13m 38s):
Yeah, yeah. Well this was kind of a funny project. I feel like it’s important to note in terms of how it was set up because you know, I’m, I’m here in Portland because it was the pandemic and we were really limited in terms of our ability to travel. The engagement that was being done. I mean, I personally wasn’t actually bleeding the engagement, the engagement was being led by folks who worked for the cities and the cohort and also kiwi bot. So the city staff and Kiwi bot staff were going out together to do engagement. And then I was meeting weekly with, to glean the learnings and hear, you know, understand what they were hearing from these engagement events. And then helping, I was helping sort of create engagement tools that they could use like surveys and things like that.

Amanda Howell (14m 19s):
So I just caveat that a little bit because it’s kind of a funny from a research standpoint to sort of be a little bit at arms length from that. But, so I think the engagement events that really had the biggest impact were truly the events. So in Pittsburgh, there were multiple times when they went to like local markets with the robots and just had them out on the closed street and allowed people to interact with them that way and ask questions. Similarly, in San Jose, they went to a lot of sort of community events with the robots and did that work in Miami? There was an or another organization involved where they had hired these community liaisons from different neighborhoods in Miami and then they had an event with the community liaisons where the community liaisons could order something for delivery with the robots and then, you know, have the experience of doing that.

Amanda Howell (15m 7s):
So a lot of it was event based. There was also just like having them out and about on the streets and you know, doing some intercept surveys where possible. So that happened more in San Jose. San Jose State University was also involved with helping do some of this data collection. So they had some urban planning students go out and do intercept surveys and do some observations of, you know, what was it like when people interacted with these devices. And so, you know, that’s, that was kind of the crux of how we did engagement. And I think, you know, by and large, I think the themes that kind of stood out the most in terms of what we heard from folks is like a lot of people said they were cute, you know, that was like a common thing.

Amanda Howell (15m 48s):
Kids tended to be really excited around them. It was like fun to see them like drive around and they have faces and it was fun to see them open and close. So I think there was some like warm and fuzzies around that. There was also a healthy amount of skepticism of like, okay, like this is fun, but like also I have a hard time trying to imagine exactly what situation I would use these, like when would I use these? A lot of people did talk about like, what does this mean for jobs? What’s the impact on jobs and delivery jobs if, if these are widely deployed? Some people express concerns around privacy, you know, especially with having more things out in the world that have cameras. Like what information is it taking in, what is it storing?

Amanda Howell (16m 29s):
So the feedback was overall, I think fairly like positive and people were open to it, but there was also like a lot of questions around it, like they wanted to know more. And I think that’s also where it was really helpful that we, you know, for most of these events there was both a city staff member and then a Kiwi bot staff member there together because the city, you know, the city staffer could speak to, like what was the goal of doing the pilot and then having the kiwi bot staffer there, they could answer questions of like, okay, this is exactly how this technology works, this is what we’re storing, this is, you know, let’s get into the nitty gritty of of the tech and really try and answer as many questions as possible in those moments. So, you know, I mean I think that that was like sort of overwhelming way of the experience with the engagement.

Amanda Howell (17m 9s):
I mean, you know, there were also, like Pittsburgh in particular, there were, especially when the pilot was first announced, there was a kind of fair amount of like not being so sure that people wanted these on city streets and feeling like, oh, we don’t want like yet another thing that we’re trying to navigate around. You know, like some people are also still feeling like I’m still trying to deal with the scooters that are deployed and taking up space. And so this is just another thing. So that broadly I would say were the overarching themes we heard in the engagement.

Jeff Wood (17m 39s):
So these are sidewalk robots. If you were to explain what one of these kiwi bots look like, you know, the kids obviously thought they were cute, but like what would you say their kind of dimensions are? What are, for folks listening that you can look it up obviously, but I wanna kind of get a feel for what size they are and how fast they go and, and what their kind of main operational standard is.

Amanda Howell (17m 59s):
Yeah, so they’re essentially the size of like a very small cooler is how I would describe it.

Jeff Wood (18m 7s):
Like an igloo cooler,

Amanda Howell (18m 8s):
Like a, a really small igloo cooler. Yeah. You know, like a picnic basket style cooler is kind of the way I would describe it. So that’s kind of about this size pro. Most folks, it probably doesn’t go a lot higher than like your mid shin, so it’s pretty low to the ground. It does have a flag that sticks up off of it so that it has a higher profile with the flag. They’re going about on average four miles an hour. So they’re not going that fast on the sidewalks. I think they have the capacity to go faster, but that’s, that’s how fast they were operating in the cities that we were deploying them on and front of, the robots have sort of little digital faces and so it can kind of make like a happy face or a sad face or put a heart on it or something so it can, you know, it’s sort of being anthropomorphized in that way.

Amanda Howell (18m 58s):
And then, you know, the way that you engage with it, so it rolls down the street slowly on the sidewalk, it crosses at intersections and then when it gets to where the location is that it’s either picking up an item or delivering an item, someone has to come out to meet the robot and then it opens like a clam shell. It just sort of opens slowly and then you can put something in it or take something out. And that’s basically how it works.

Jeff Wood (19m 24s):
It seems like it can hold like a sandwich or maybe an individual meal, right? Like that’s kind of the size.

Amanda Howell (19m 29s):
Yeah, it’s pretty small. That’s right. I mean I think like you could probably have, you know, depends on what you’re ordering. You could probably have a couple of sandwiches delivered, maybe like two kind of tall lattes. It’s capacity is pretty limited.

Jeff Wood (19m 44s):
What does that mean for the companies that the pilot was trying to partner with? Were they like, Yeah sure we, we wanna deliver sandwiches in in these vehicles or Mexican food or whatever it might be. What were the feelings like from the folks that were you were trying to partner with in order to do some real world testing?

Amanda Howell (19m 59s):
That is a great question and that is definitely one of the, I think, biggest challenges that we came across over the course of the pilot. So Kiwi, but approached a lot of local stores and restaurants in these four cities. And I think that, you know, people were interested, folks were like, Oh yeah, this like, it’d be interesting to try this technology, but at the end of the day, from the standpoint of rubber meeting the road, how does this actually work? There were quite a lot of challenges in trying to figure out how this would integrate into a business, especially with also just, you know, I think with what everyone was dealing with too at the time, like limited staff and you know, covid protocols and all kinds of stuff. It was, I mean there was already so much going on.

Amanda Howell (20m 40s):
And then on top of that, some of these places they did end up working with a taco truck in Pittsburgh, but that truck didn’t have an online menu to begin with, you know, so Kiwi bought, ended up then having to help digitize their menu and then help create like an ordering platform, like things that didn’t actually even exist before. So like that I think was the most kind of extreme versions of some of these challenges that in terms of trying to integrate with these businesses made it pretty difficult. And then also like, you know, sort inventory management is another thing. Like if businesses aren’t necessarily used to doing a lot of online orders, and again, like I think everyone was like figuring this out during the pandemic and, and scaling up and doing a lot more of it.

Amanda Howell (21m 20s):
But you know, even so like some places were like, Yeah, that seems cool, but like that really isn’t where my business is. You know, people call and they place an order and they come get it. And so, you know, I don’t know that like there’s gonna be the demand for it from my customer base. So, you know, there were quite a lot of challenges in that regard in terms of actually like kind of onboarding, especially the small and local businesses, which is I think one of the things that we had really wanted to try, which you know, was to see like what is it like to try and work with these smaller local businesses? Could this be like from an economic development standpoint, can this be beneficial? And you know, I think the challenges of onboarding and then, you know, moving forward for the folks who did actually end up using it, it’s like it was a fun novelty, but it’s going four miles an hour.

Amanda Howell (22m 3s):
Like that’s pretty slow. Yeah. So I think a lot of places were also just like, well, you know, we’re kind of already using these, those are delivery services and that’s working fine. Yeah, maybe it’s imperfect, but this is what we’ve got.

Jeff Wood (22m 16s):
There was like a distance limit too, right? Like a mile and mile and a half as well for you. All

Amanda Howell (22m 20s):
Right? Yes. The distance was basically between a mile and a half to two mile radius delivery. So that’s also a challenge. I think you’re kind of getting at a lot of the questions around like what are the really applicable use cases of this technology right now, given the limitations. You also, I think in that limited or radius, you need a pretty high density of demand and that doesn’t exist in a ton of places, you know? Yeah. Where you have a lot of places that are both, like you have local restaurants and, and goods and services and things, and then a lot of people living within that same small radius that are gonna order that delivery. So, you know, I think that’s, that raises questions like, okay, well what are the sustainable use cases of this technology right now?

Jeff Wood (23m 3s):
Do you think that doing the pilots kind of told you that small businesses really aren’t the target for some of these robots and maybe it’s something else?

Amanda Howell (23m 9s):
I think as it currently stands, I would say yes, I, I think right now it does not feel like that’s the target for this technology and, and I’m not sure that there’s like a huge value add from the business standpoint currently to use these devices for delivery. Especially when you’re having people deliver on, well certainly by car, by bike, by foot, you know, if they’re going as fast or faster, that’s probably not the best use of the technology right now.

Jeff Wood (23m 35s):
Another thing is like, you know, they’re called personal delivery devices, PDDs, which is kind of a wordy way to say little robot or sidewalk, robot. I don’t know, is there, is there a way that you like PDD or a sidewalk robot or what’s the term dejour? Like what’s the best one to use?

Amanda Howell (23m 48s):
I mean I think sidewalk delivery robot is the thing that actually makes sense, right? It’s like okay,

Jeff Wood (23m 53s):
Like that’s what it is.

Amanda Howell (23m 54s):
That’s what it is, right? Like if I don’t work in this space, like a personal delivery device is like that is strange. You know, I think most people are like, I don’t, what does that mean? Yeah. And so like a sidewalk delivery robot. Yeah, that makes sense.

Jeff Wood (24m 6s):
Yeah, I think the interesting thing is that there was kind of a a thinking about a pivot where you know, there might be some value in these robots not just from deliveries but from actually like mapping the infrastructure of a neighborhood because of the challenges that they encounter. Kind of the similar challenges that pedestrians and people with disabilities and wheelchairs and others get all the time when they’re trying to navigate a city. They even got hit by cars apparently a couple times because they couldn’t see them taking a right turn, which there’s a whole discussion nationally now after Washington DC just banned right turns on red. So there’s all these things that maybe you learned that they could do that maybe wasn’t a used case before.

Amanda Howell (24m 42s):
Yeah, absolutely. So that definitely was a big pivot. Basically I think I would say midway through a lot of the pilots, which was that okay, this idea that we could maybe serve small and local businesses and provide them with support in this difficult time, Okay, well maybe this is actually not gonna work so well. But you know, these devices are going to all these places and they’re low to the ground and they have the opportunity, they have all this technology, like they do have the opportunity to collect infrastructure data that could be really valuable to cities to really identify like where are the cracks in the sidewalk and you know, where is the sidewalk misaligned or where are obstructions? Where do those exist? And so that definitely sort of ended up being a direction we started to try and move towards kind of mid pilot along with Kiwi bot of like, okay, well what would it look like to use these devices for this purpose?

Amanda Howell (25m 33s):
And then it got into like interesting questions around like, well this isn’t actually the area that Kiwi bot had originally gotten into, you know, this was, this wasn’t what they had created the robots for. So then sort of trying to figure out, okay, well what does it take to actually collect the types of data that is really valuable to cities? What is that data? How do we make sure that these two things kind of speak together, that they’re collecting it in a way that’s actually usable by engineers and city staff? And like all of that kind of ended up then getting like really interesting questions but then it’s own challenges, right? Of like, okay, like that was not there of expertise. And so I think like kind of trying to figure out like how do we have these folks who are kind of in some ways like sort of for lack of a better way of thinking about it, like speaking a different language where you’re like, now we’re talking like an infrastructure language and there’s like the really nitty gray of like the details of exactly what the measurements that need to be taken in and stuff like that.

Amanda Howell (26m 24s):
And can Kiwi bot sort of move towards doing this in a way that’s actually gonna be usable? So I think that’s an interesting question. You know, we kind of ended the pilot with that being where the direction that Kiwi bought was trying to take it in. Ultimately we weren’t deploying for long enough and able to get to the point where we were collecting data that ended up really being usable by the cities. I think that’s something that, a direction that they are continuing to explore, you know? But then I think it becomes a question of like, okay, well so does a robot, does the way it built need to change? Like if it’s gonna actually collect infrastructure data but not do deliveries, doesn’t need an insulated cooler, Like how does the, how does the design of it change as a result?

Amanda Howell (27m 7s):
Because now it’s doing something entirely different.

Jeff Wood (27m 9s):
Yeah, I imagine you could put Lidar on it and then you need all of the memory in that space, or you need an accelerometer that collects when it bumps and stuff like that and those types of things. And they take up space, especially in those vehicles that are driving around San Francisco here, I can tell that they take up space.

Amanda Howell (27m 26s):
Yes, definitely.

Jeff Wood (27m 28s):
How was the pilot beneficial to the local staff, the city staff? I mean obviously there’s a way for you to take information and then use it because it was such a nude technology and a new idea that you can use it to write regulations, but were there other ways that the local staff benefited from the pilots?

Amanda Howell (27m 44s):
I would say I, I think yes, absolutely. Just the opportunity to work so closely with a tech provider and you know, get out in the field together and do that engagement was beneficial. And I think, you know, we kind of talked about this a little bit in the report, but I think from the standpoint of when we started the pilots, there was this sense of like, oh my gosh, like these are coming so fast, we need to figure it out. We need to have a pilot program. They’re gonna be everywhere tomorrow. And I think the opportunity to really dig in and learn about the tech and try deploying it in different circumstances allowed everyone to sort of see up close like, oh actually, you know, this is an opportunity but this is a big challenge and this challenge is gonna take some time to figure out.

Amanda Howell (28m 28s):
So like, this actually isn’t maybe as pressing as I felt like it was. And so there’s more time to continue to iterate to figure out like what is the best way to deploy the technology. I think it gave him a sense of being able to kind of take a deep breath around some of this stuff, you know, because I think that with tech, you know, I do think that sometimes it can just feel like, oh my god, the walls are closing in and we have to figure this out. So I think that that was, I think that was a big benefit. I think it was kind of an unusual pilot in the idea of trying to really co-create these use cases with a, with a tech provider. And I think that was really a beneficial thing for everyone involved of what does this look like to try and do this kind of partnership? How can we do this again and what are we learning from it to continue to make it better?

Jeff Wood (29m 11s):
What’s interesting to me is that, you know, we have all these other deployments of technology in the past for transportation and they’ve gotten kind of mixed reviews or frustrated cities. So one of my questions is, were there lessons from this pilot project that you feel like the deployments of like say Uber or a bird or somebody along those lines could have learned from, if they would’ve, you know, done something like this rather than just throw everything on the street and you know, shoot first, ask questions later?

Amanda Howell (29m 35s):
Yeah, well I mean I think it comes back to and to sort of, I think be reductionist like building that trust. And so I do think that learning from this and the way that this pilot was structured allowed the cities and the tech provider to start to build trust at an earlier stage and try and start to build these relationships earlier on in the like testing and deployment of the tech. You know, I mean I think that this has been the ongoing challenge and conversation and discussion around, you know, Uber and Lyft and Micro mobility and ECUs and all this kind of stuff is this sort of sense that I think a lot of city staff end up just kind of feeling like they’re getting caught on their heels and then everybody always feels like they’re trying to play catch.

Amanda Howell (30m 21s):
And I think that partnering earlier on, allowing staff to really see what the technology can do and is capable of, I hope that down the line as we continue to see it develop, it allows for more thoughtful piloting and potentially kind of less antagonistic relationships where it feels like, okay, like this new technology, I mean Uber and Lyft Mobility, e scooters, all this stuff, like these things are not all or nothing like it’s not, it’s not all bad, it’s not all good. There are good things that have come of it. There are negative extra externalities. But I think that the way that those other services in particular were initially deployed, I think sort of ended up creating this environment where it just sort of feels like the cities are, you know, just trying to play catch up and then it just becomes sort of about regulation and then there’s like this like dynamic of like, we don’t like the regulation and we don’t want this and blah, blah.

Amanda Howell (31m 16s):
I just think like trying to do this sort of private public partnership piloting earlier on will hopefully allow for a better service in the future that hopefully does serve more people. That’s the goal.

Jeff Wood (31m 29s):
Yeah. So there were three main objectives for the pilot. Learn about the technology and potential issues, educate and engage the public and then work with private sector companies. Looking back, which objective brought you the most interesting findings?

Amanda Howell (31m 42s):
Well, I think they all yielded interesting findings. I do think number one ended up just being, which was learning about the technology. There was so much that we learned about the tech by doing it, doing this kind of deployment, just in terms of like things that I think we couldn’t have anticipated otherwise. So for instance, in Detroit, Kiwi bot had partnered with a restaurant that then in order to go to the kind of main deployment area, the neighborhood of Corktown, which is where they were gonna be doing deployment, the robot would have to cross Michigan Avenue, which is overseen by the state, so Michigan, d o t, you know, it’s under their jurisdiction. And the, the robot could not get across the street before the light changed.

Amanda Howell (32m 27s):
So then it would kind of end up looking like it was running a red light because it couldn’t actually get all the way across. That’s something that we hadn’t really thought of in advance of just like, that’s really a limitation of our infrastructure, not the technology, but yeah, and how we’ve designed that. But I think that was a thing where then ultimately the way that we ended up having to deal with that or they ended up sort of saying like, okay, well this is, you know, this is becoming an an issue so we’re just gonna end up having to create a different path and so we’re gonna have to avoid Michigan, AOL together and we’re gonna take this like even more circuitous route right to delivery. I mean, I think kind of learning about the technology and the interactions within the environment, our built environment as it exists, I think was just absolutely fascinating.

Amanda Howell (33m 10s):
And I think that with the previous deployments primarily being on college campuses, like you just didn’t really get to figure that stuff out before because those campuses are, they’re smaller infrastructure conditions tend to be sort of better, at least more consistent. And then again, like in Pittsburgh, right? It was like there were a lot of neighborhoods where the robots just couldn’t even navigate because like there’s like all these overgrown bushes and all these things are like, wow, they can’t, they couldn’t carry on the sidewalk. So, you know, I mean I think things like that. And then just also like in terms of weather conditions, what can they operate in and what can they not? I think that yielded really interesting learnings, being able to just really learn about the tech and see what it could do.

Amanda Howell (33m 51s):
And again, learn about the tech in relationship to our build environment.

Jeff Wood (33m 55s):
Yeah, I feel like humans, when they get into that situation where they’re forced to cross the street faster, you, you run across the street, even though you’re a pedestrian, you have the right of way, you’re like scared of the vehicles. So what does that mean for a robot? Like they’re not really scared, but they might get crunched and then can you speed up, can you slow down? Like where would you go? It is all kinds of interesting things happen when you learn about how, you know, our infrastructure is failing pedestrians and cyclists in many different ways and, and obviously it does the same for, you know, sidewalk robots. Yep. So what’s next for you all? Like what’s the next step? Is there another pilot coming? Is there more research about these? PDDs is something else coming down the line?

Amanda Howell (34m 34s):
So the Night Foundation, the night AV initiative is ongoing through the end of 2023. We’re not doing anything at this stage, anything more with delivery robots. That phase has ended really now the cities that we’re working with have kind of moved into, they’re doing different pilots and engagement work now. That one was a really unique and I think awesome opportunity for all four of them to do the same thing at once, which I mean had a ton of value. And I will just make a plug in terms of like, I think that’s a great way to the extent that some cities can work together and deploy something at the same time. There’s a ton of learnings that can happen that way. So now the cities are doing different work. Miami Dade currently has a, an autonomous shuttle running, it’s at the zoo, so they’re testing that out.

Amanda Howell (35m 18s):
Detroit is also working towards doing an autonomous shuttle deployment now. So the initiative is sort of moving in different directions, but ultimately I think like the goal and the something that we’re all continuing to work towards together is definitely sort of how can we continue to best prepare cities for further autonomous deployment. And so we as urban as a next, and then in partnership with Citi five, who’s also part of the night a initiative, we’re working on an autonomous vehicle playbook that we’re putting out next year that, you know, kind of pulls together a bunch of learnings from this multi-year effort of like, what does all those look like and, and what do we learn from all these different engagement opportunities and things that have happened over the course of the initiative.

Amanda Howell (36m 2s):
So that’s what we’re working on now.

Jeff Wood (36m 3s):
Awesome. Well, so folks wanna get a copy of the report or find out more about Urbanism Next, where can they do that?

Amanda Howell (36m 9s):
Our website, urbanism next.org has all of our reports, including the most recent Sidewalk delivery robot report and information about Urbanism Next and the work that we do. And I will also note that we have an annual conference that our next conference is coming up next April in 2023. So yeah, please, I would encourage folks to check it out.

Jeff Wood (36m 28s):
Awesome. Well, Amanda, thanks for joining us. We really appreciate your time.

Amanda Howell (36m 31s):
Thank you so much for having me. I really enjoyed chatting about this

Listen to the Talking Headways Podcast


…the first thing I read every morning is the newsletter to see what’s been out there. It’s great to have an aggregator that pulls everything together so nicely.

Joe Cortright, City Observatory

I think that the email newsletter that you do every morning is the best one that I get, and I get a lot of them.

Mary Newsom, The UNC Charlotte Urban Institute

Really is the best daily urban newsletter out there.

Eric Jaffe, Editorial Director Sidewalk Labs


To Receive The Overhead Wire in Your Inbox Daily

Premium Daily Subscription

The Premium Daily Subscription is our most information packed offering, chock full of over 30 pieces of news every single day. Included are popular features such as the quote of the day and the most read article from the previous day. Also included is our weekly roundup for times when you are strapped for time but need to know what’s going on.

Premium Weekly Subscription

The Premium Weekly Subscription is for professionals constantly under a time crunch. We take the most read items from the week before and share them with subscribers along with more in depth analysis of a single popular topic.

Learn More and Subscribe

Video of the Day

Friends of The Overhead Wire

Back To Top

Welcome to The Overhead Wire

What Can We Help You Find?

Try Our Newsletter For Free