Thinking about Training

1 Refer to the weather forecast (right) in your response. During a week of daytime test drives:

  • On Wednesday and Friday, the self-driving car drives safely.

  • On the other days, the self-driving car is unsafe.

What might explain why the car drove unsafely on some days?

a one-week weather forecast predicting a variety of different weather conditions

2 Imagine that a self-driving car has done extensive training on a one-lane road, in all weather conditions, and at all times of day. Would you expect it to be able to safely drive on a busy two-lane road? Explain.

3 Imagine that a self-driving car trained on isolated country roads, as well as city streets, and highways in all weather conditions and times of day. It would be pretty safe on many roads! What might it still not be prepared to navigate safely?

Confidence Rating

4 In addition to producing a steering angle for each image of the road, ALVINN produces a numeric measure of "confidence" in its response.

  • What do you think causes ALVINN’s "confidence" to increase or decrease?

  • Is 100% "confidence" possible?

These materials were developed partly through support of the National Science Foundation, (awards 1042210, 1535276, 1648684, 1738598, 2031479, and 1501927). CCbadge Bootstrap by the Bootstrap Community is licensed under a Creative Commons 4.0 Unported License. This license does not grant permission to run training or professional development. Offering training or professional development with materials substantially derived from Bootstrap must be approved in writing by a Bootstrap Director. Permissions beyond the scope of this license, such as to run training, may be available by contacting contact@BootstrapWorld.org.