Today we muse about slow and beautiful travel across North America. We marvel at the creativity of toll evaders and those who catch them. And, we see where self-driving cars go to study road and traffic signs. It is a lot to think about and enjoy.

The most stunningly beautiful train rides in America

For many, the best way to travel is by train. There are almost none of the security lines one faces at airports, weather is not a major problem, seats are big, stretching one’s legs is easy, and the passing countryside is soothing. Plus, it is a time of relaxation that many of us in the modern age never get. Here is a collection of the slower way to travel, by rails across North America.

  1. Cascades — Eugene, Oregon, to Vancouver, British Columbia; Distance: 156 miles; Cost per ticket: $90
  2. Cass Scenic Railroad — Cass, West Virginia; Distance: Eight miles; Cost per ticket: $37
  3. California Zephyr — Chicago, Illinois, to Emeryville, California; Distance: 2,438 miles; Cost per ticket: $167
  4. White Pass & Yukon Route — Skagway, Alaska; Distance: 120 miles round-trip; Cost per ticket: $122-$229
  5. Capitol Corridor — San Jose to Auburn, California; Distance: 168 miles; Cost per ticket: $43
  6. Cape Cod Central Railroad — Hyannis to Buzzards Bay, Massachusetts; Distance: 27 miles; Cost per ticket: $22-$47
  7. Grand Canyon Railway — Williams, Arizona; Distance: 130 miles round trip; Cost per ticket: $79
  8. Great Smoky Mountains Railroad — Bryson City, North Carolina; Distance: 32 or 44 miles; Cost per ticket: $51-$56
  9. Coast Starlight — Los Angeles, California, to Seattle, Washington; Distance: 1,377 miles; Cost per ticket: $120
  10. Sunset Limited — New Orleans, Louisiana, to Los Angeles, California; Distance: 1,995 miles; Cost per ticket: $134
  11. Strasburg Rail Road — Strasburg, Pennsylvania; Distance: Nine miles; Cost per ticket: $15
  12. Empire Builder — Chicago, Illinois to Seattle, Washington; Distance: 2,206 miles; Cost per ticket: $179

Follow that car! Toll-collecting bureaus step up enforcement

Catching toll evaders has become a game for authorities. From New York to California and Minneapolis to Dallas, toll collectors are amassing a big bag of tricks to foil evaders. The evasion tricks are amazing. They include, “truck driver accused of using fishing line to flip his license plate to avoid capture, or the motorcyclist who used a toggle switch to retract his plate.” And, other evaders are costing tunnel and bridge authorities millions of dollars every year. The rest of us end up paying.

Detecting and catching violators has become a high-stakes game of cat-and-mouse.

A truck driver was accused in March 2016 of using fishing line to flip his license plate to avoid paying tolls from New Jersey into New York City on the George Washington Bridge. Police found a fishing line rigged from the cab to a hinge on the front license plate. They say the line could flip the plate out of view going through the toll plaza, while the rear license plate was bent up to defeat security cameras.

A few months later, a man on a motorcycle used a retractable license plate to skip a toll at the Holland Tunnel in Jersey City. Authorities say the license plate was concealed as the motorcycle entered an E-ZPass lane, and an officer then saw the man use a toggle switch to return the license plate to its proper position.

The motorcyclist eventually pleaded guilty to a disorderly persons offense in municipal court.

Click here to subscribe
Teaching cars traffic signs

With the coming of driverless cars, one of the main efforts is programming the car’s computer to read signage. What does a Polar Bear Crossing look like to a computer? How about “Falling Rocks”? And, in New Hampshire a driverless car will have to look out for “Moose.”

Some robotic vehicles are learning signs from photographs of real-world proving grounds. Others are “studying” global signage in their labs, using images from one of the world’s largest databases of street-view photographs. That’s what the Swedish startup Mapillary, a crowdsourced alternative to Google Maps’ Street View, is up to. It’s using machine-learning technology to sort through its 114 million ground-level images —spanning 1.6 million miles of streets — and pluck out features most relevant to robotic cars, which it then sells in neat data-sets to AV manufacturers and tech companies. Right now, the focus is on Traffic Signs 101.

Autonomous vehicle software can be “trained” to read thousands of signs in Mapillary’s library of 500-odd sign types from more than 60 countries, using what engineers call “neural networks” — a computer system that “studies” lots of tweaked versions of a thing to learn how to infer what it is, at any angle or condition. The more traffic signs the software sees, the better it can commit to memory the difference between, say, an underpass and a train crossing, and how those might be represented in Japan versus Norway versus Nicaragua. Then the car can respond accordingly in the real world.

Photo: White Pass & Yukon Route