Guidelines for Safe On-Road Testing of SAE Level 3, 4, and 5 Prototype Automated Driving Systems (ADS)
This document provides guidelines for the safe conduct of on-road tests of vehicles equipped with prototype conditional, high, and full (levels 3-5) automated driving systems (ADSs), as defined by SAE J3016. It does not include guidance for testing production ADSs intended for sale to the general public. The scope is further limited to testing of automated prototype vehicles on public roads.
These guidelines do not address:
Testing of driver assist (Level 1) or partial (Level 2) automation systems, which rely on a human driver to monitor the environment. (See SAE J3016 for Definitions of Levels of Automated Systems.)
The precise regime of road testing for a particular prototype will depend on the intended level of automation and the targeted capabilities of the prototype (see SAE J3016 for more information).
A prototype suitable for testing on public roadways is presumed to have already passed laboratory and/or closed-course testing, which are not addressed by this document.
This document provides general safety-relevant guidelines for performing tests of prototype automated driving systems (ADSs) equipped on test vehicles operated in mixed-traffic environments on public roads. The levels of automation addressed in this document include conditional (level 3), high (level 4), and full (level 5) as defined by SAE J3016; when activated, these ADSs do not rely on a human driver for monitoring and responding to the vehicle or traffic environment. (SAE J3016 defines the italicized terms in this document.) These guidelines apply to the testing of all types of motor vehicles including light-duty, passenger, freight or transit vehicles, but are not concerned with component-level testing. These guidelines address only the safety-related prerequisites for on-road testing and conduct of such tests. They do NOT establish performance criteria or test procedures for production vehicles equipped with ADSs.
Mixed traffic environments are active public roadways, closed campuses (such as military bases, factories, ports, and enclosed communities), and other contexts that involve a risk of injury to road users or damage to their property. These environments will (variously) include motor vehicles of all types and classes, pedestrians, and pedal cyclists, as well as animal and object hazards. Accordingly, these guidelines address analytic, laboratory, simulation, or closed-course test methods only to the extent that these methods are part of, or predicates to, tests of vehicles equipped with prototype ADSs conducted in mixed-traffic environments on public roadways. The guidelines assume that any such tests will be conducted under some form of human supervision; this assumption applies even if the ADS being tested is considered to be a level 5 “full automation” system. Licensing and registration requirements for test vehicles equipped with prototype ADSs should be checked with relevant local jurisdictions. This document also provides guidance on test driver training and test program management; graduated road testing; test data capture; safety override guidelines, and software development and release requirements.
Safety guidelines for the on-road testing of vehicles equipped with prototype ADSs do not currently exist. The information given in this report will assist interested parties in the creation of guidelines for the on-road testing of such prototypes.