SUNRISE Safety Assurance Framework

In Use Case 1.1, the SUNRISE SAF is tested for perception systems in urban environments. The aim here is to test those systems in complex urban intersections/scenarios and adverse weather conditions. There are camera-based, LiDAR-based and radar-based perception systems.

There are six behaviours defined for UC1.1: Car-following, outdoor parking with other vehicles entering or leaving parking spots, obstacle avoidance due to construction work, pedestrian and bicycles, and mainly urban intersections like roundabouts and right/left turn junctions. All of them with day/night conditions and/or adverse weather conditions like fog, rain and others.

In Use Case 1.2, the SUNRISE SAF is tested for cooperative perception and decision-making in urban intersection scenarios. The system under test is a cooperative Adaptive Cruise Control (ACC) that is expected to improve with enhanced perception obtained through Vehicle to Everything (V2X) connectivity.

There are four functional scenarios defined in T7.1. Three of these scenarios involve improving vehicle behaviour when approaching a traffic light and receiving traffic light phases, timings, and map information. The scenarios cover three types of unexpected events: a baseline scenario with no unexpected event, a violation by a distracted pedestrian, and a reset of phases due to a pedestrian call. The fourth scenario deals with a red-light violation by a crossing car, and in this case, V2V information complements the traffic light connection.

In Use Case 1.3, the SUNRISE SAF is tested for cooperative perception and decision-making in urban intersection scenarios. The SUT is the perception AD subsystem of an urban chauffer ADS which is also capable of sending/receiving and processing (on-board or off-board e.g. employing a remote smart RSU node) rich V2X information data mainly consisting of object-level data perceived from other road users.

There are three functional scenarios defined from T7.1 for assessing the cooperative perception system performance in Use Case 1.3. The first is the “darting-out pedestrian” scenario (1.3 A), the second one is the “urban junction” scenario (1.3 B), and the third one is the “urban roundabout” scenario (1.3 C).

Use Case 2.1 involves testing the SUNRISE SAF with an AD function that controls the ego-vehicle in a traffic jam scenario. The system is expected to control both its longitudinal and lateral position by a combination of braking, and steering commands. All of this is expected to happen while respecting a few constraints, such as a maximum longitudinal speed and a distance to the vehicle in front of the subject vehicle.

Use Case 3.1 involves testing the SUNRISE SAF with map-based AD functions for highway scenarios. The systems under test will include an advanced ACC/HWP designed to control longitudinal dynamics based on HD map data and onboard sensors.

There are two defined functional scenarios: adapting speed to varying speed limits and varying road curvature, using information from sensors and maps. [NOTE: A third functional scenario, green driving on slopes, initially proposed in SUNRISE deliverable D7.1, has been dismissed.]

The main aim of sub-UC 3.2 will be to demonstrate how safety could be improved on motorways by including cooperative functions in the HWP system. One example of this HWP functionality is the leveraging and upgrading of the driver assistance functionality developed previously in C-ACC from sub-UC 1.2.

By applying the SUNRISE SAF, the ODD coverage can be demonstrated, not spatially or functionally to specific operating fields and conditions. This is possible by taking advantage of different virtual validations, e.g., usage of extended realistic scenarios where cooperative manoeuvres between agents can be proven by variating all the interesting parameters, even in corner cases or in safety-critical conditions. After extensive virtual tests demonstrating the virtual ODD coverage, the designed cooperative functions can be proven in real-world scenarios regarding the main identified use cases and parameters, in order to provide a sufficient test coverage of the ODD.

Confined areas, with perimeter protections and a reduced risk of unauthorised entry, offer the advantage of a well-defined ODD. These environments and highway use cases are expected to be among the first to support deploying highly automated (L4) heavy vehicle functions, aspects that make them essential to SAF validation efforts.

The reverse parking of a truck with a semitrailer is a strong use case for validating specific aspects of the SAF due to the controlled environment and limited range of possible scenarios. Validation in such spaces benefits from lower speeds and regulated traffic conditions, although the presence of mixed traffic and humans introduces additional complexity.

UC4.1 focuses on the reverse parking manoeuvre of a truck with a semitrailer. The idea from the application phase is that the sub-UC is a reverse-driving truck with a semitrailer at a logistic hub.

In Use Case 4.2, the SUNRISE SAF is tested for truck low-speed connected perception cyber-testing. The SUT is a connected perception AD subsystem that is compromised by cyber-security threats. The main aim is to combine in a simulation environment several aspects simultaneously (physical environment, perception, V2X connectivity) and study the effects of physical or remotely executed cyber threats on collective environment awareness.

There are two functional scenarios defined from T7.1 for assessing the connected perception cyber-security performance in use case 4.2. The first is the “distorted camera input” scenario, whereas the second one is the “CPM message attacked” scenario.