AV Accident Analysis from CA DMV - PowerPoint Presentation

AV Accident Analysis from CA DMV Nazanin Nader – Industrial and Systems Engineering

Autonomous Vehicles and Level Definition 2 Reproduced AS-IS with permission from SAE-International J3016 TM T3e Webinar – Talking Technology and Transportation in Education

Autonomous Vehicles and Level Definition 3 Reproduced AS-IS with permission from SAE-International J3016 TM T3e Webinar – Talking Technology and Transportation in Education

Where are we now? Why is it important? 4 T3e Webinar – Talking Technology and Transportation in Education

Where are we now? Why is it important? 5 T he human driver is responsible for monitoring the safe operation of the vehicle at all times, and must be capable of taking over immediate control in the event of an autonomous technology failure or other emergency CA DMV “Autonomous Vehicles Deployment Regulation” T3e Webinar – Talking Technology and Transportation in Education

Accidents’ Reporters Overview 6 37 manufacturers have permits to test on public roads (growing list) 5 of them reported accidents (collisions and/or injuries involved) up to July '17 T3e Webinar – Talking Technology and Transportation in Education

Accidents’ Reporters Overview 7 T3e Webinar – Talking Technology and Transportation in Education 37 manufacturers have permits to test on public roads (growing list) 5 of them reported accidents (collisions and/or injuries involved) up to July '17 26 accidents up to July 2017

Accidents’ Reporters Overview 8 Autonomous miles driven T3e Webinar – Talking Technology and Transportation in Education 37 manufacturers have permits to test on public roads (growing list) 5 of them reported accidents (collisions and/or injuries involved) up to July '17 26 accidents up to July 2017

Accidents’ Reporters Overview 9 Autonomous miles driven 1,060,199 Google T3e Webinar – Talking Technology and Transportation in Education 37 manufacturers have permits to test on public roads (growing list) 5 of them reported accidents (collisions and/or injuries involved) up to July '17 26 accidents up to July 2017

Accident Overview 10 T3e Webinar – Talking Technology and Transportation in Education

Accident Overview 11 7 9 24 T3e Webinar – Talking Technology and Transportation in Education

Detection of Rear-End Collisions 12 62% of accidents are low speed rear-end collisions with the AV in front Conventional vehicles  30% of the accidents are rear-end T3e Webinar – Talking Technology and Transportation in Education

Detection of Rear-End Collisions 13 62% of accidents are low speed rear-end collisions with the AV in front Conventional vehicles  30% of the accidents are rear-end An AV in the rear will deploy automatic breaking to avoid a rear-end collision Can an AV in the front detect a collision? T3e Webinar – Talking Technology and Transportation in Education

Detection of Rear-End Collisions 14 62% of accidents are low speed rear-end collisions with the AV in front Conventional vehicles  30% of the accidents are rear-end An AV in the rear will deploy automatic breaking to avoid a rear-end collision Can an AV in the front detect a collision? T3e Webinar – Talking Technology and Transportation in Education

Vehicle Make and Appearance 15 Are rear-end collisions a problem of conventional drivers’ reaction to AVs? T3e Webinar – Talking Technology and Transportation in Education

Vehicle Make and Appearance 16 Type of Vehicle Total number of Accidents Total Miles Travelled Accident Frequency Google Prototype 10 403,226 2.4e-5 Retrofitted Lexus 12 649,841 1.8e-5 Are rear-end collisions a problem of conventional drivers’ reaction to AVs? 2.4e-5 1.8e-5 2.8e-4 1.2e-4 5e-5 T3e Webinar – Talking Technology and Transportation in Education

Correlation with Mileage Driven 17 Correlation between cumulative accidents and cumulative autonomous miles The plot shows that accidents are still happening  normal trend for conventional vehicles, but is this going to change as AV technology is further developed? Gaps and stationary trends are expected in the future as fewer accidents are witnesses (but mileage travelled keeps increasing) This is due to the technology involved: AV “brains” use Machine Learning: as the algorithms are tuned and learn to handle more scenarios we expect a lower number of accidents T3e Webinar – Talking Technology and Transportation in Education

Accidents Trends after July 2017 18 T3e Webinar – Talking Technology and Transportation in Education The second half of 2017 saw 28 reported accidents of AVs 50% of those are rear-end accidents, confirming this is the current problem to tackle The majority of the accidents for the second half of 2017 are reported by GM, with Google having moved the majority of its testing to other States. rear-end breakdown

Preliminary Conclusions - 1 19 Predominance of rear-end accidents Problems of detecting rear-end accidents by both driver and AV No demonstrated merit to the claim that people purposefully collide with AVs No demonstrated merit to the claim that anthropomorphous vehicles will have a lower frequency of collisions, i.e., be safer We analyzed accident that occurred on public CA roads Semi- Autonomous vehicles require the collaboration of human drivers  HMI Our analysis shows preliminary issues that need to be addressed in the HMI realm: T3e Webinar – Talking Technology and Transportation in Education