The tragic accident that saw an autonomous Uber vehicle strike and kill a pedestrian has led many to call for a dramatic slowdown in testing and increased regulations.

Some engineers believe such a slowdown makes the most sense in the long run and will ultimately lead to better autonomous cars with fewer flaws.

One of those is Aviral Shrivastava, an associate professor of computer science at ASU. He teaches an embedded computing course that includes involving students engineering a self-navigating toy race car, and he thinks there is a safer, more cautious approach to testing the technology. He joins The Show's Steve Goldstein to talk about it.