Generally, I think this is an excellent presentation of an undervalued mobility resource. As an autonomous vehicle advocate, however, I do feel compelled to make a few arguments regarding the future of walkable space, as my contention is that we could get the outcome you're looking for without the need for specific action or investment:
A. Just to get this out of the way: while no one can say with certainty when self-driving vehicles will begin commercial roll-out, there are many (myself included) who see 2025 as a very reasonable timeline for significant penetration of shuttles, taxis, etc. in areas where human-driven cars would be excluded. If you'll humor that timeline, the obvious follow-up question is: how long will it take to construct prominent sidewalk installs/upgrades, and how long would those sidewalks need to be operational to deliver ROI if the advent of driverless traffic made them unnecessary?
B. I ask because there are several inherent design components of autonomous vehicles (AVs) which imply (again, humor me and everyone else who subscribes to these assumptions) that walkable space will be created organically, without the need for major construction: - AVs cut down on the total number of vehicles in operation, due to ridesharing options and a usage-based model rather than ownership-based. Fewer vehicles and fewer parking spaces = more space for people. - AVs minimize the need for convenience-based parking, which means the ~50% of a street's pavement currently relegated to parked cars could become the sidewalks while the cars park themselves in a garage or lot somewhere. - Having fewer cars, and having many of them off the streets when parked, is how you get your space... but how you ensure the space feels safe is by way of the AVs' driving behaviors, which not only aim to be safer and more considerate of pedestrians by design, but also drive slower, more predictably, and (one assumes) with a standardized communication language between vehicle and pedestrian. In short, moving vehicles in the future will not be feared by pedestrians. - While AVs certainly don't need to be electric-powered, there are few outside of traditional carmakers who believe they'll be anything but. To that end, the picture we've painted now is not only a significant increase in walking space and a significant decrease in fear/danger of traffic, but also a significant (almost total) decrease in the health risks and unappealing experience of being near gas-powered vehicles. This on-road ped space could be designed to house seating, gardens, playgrounds... all manner of amenities that would be frowned upon with conventional traffic's pollutants.
Long story short: if you believe the sidewalk project is a 2-3 year investment, and you believe the proliferation of self-driving tech is 2-3 decades away, then your proposal seems solid. If instead, you believe the sidewalk project is a 5-year effort, and you believe self-driving tech will have a significant presence by 2025, then I would contend there is a lot more usable pedestrian space coming just around the corner.