Construction of a cosimulation experimental platform for low-speed emergency braking

Low-speed driving scenarios, such as those in underground parking lots, campus roads, factory roads, residential communities, and narrow passages, are characterized by short vehicle–obstacle distances, frequent stop-and-go maneuvers, and limited driver reaction time. Although crash severity in these scenarios is generally lower than that in high-speed driving scenarios, minor collisions, rear-end crashes, and obstacle-contact accidents still occur frequently due to delayed braking responses. Therefore, low-speed emergency braking is an important active safety function for intelligent vehicles. However, real-vehicle testing of this function is costly and risky, and it is difficult to repeatedly reproduce identical low-speed scenarios under controlled conditions. To support early-stage function verification, parameter debugging, experimental teaching, and reusable test case design, this study aims to develop a function-oriented co-simulation experimental platform for low-speed emergency braking using CANoe and MATLAB/Simulink.

The proposed platform adopts a four-layer architecture consisting of a scenario input layer, a model computation layer, a communication interaction layer, and a result output layer. In the CANoe environment, system variables and a visual Panel interface are configured to enable the setting of key test parameters, namely, target vehicle speed, obstacle distance, safety distance threshold, vehicle load, road adhesion coefficient, and braking enable state. In the MATLAB/Simulink environment, a vehicle longitudinal dynamics model, a braking distance estimation model, and a low-speed emergency braking logic model are established. The longitudinal dynamics model considers the effects of driving force, rolling resistance, aerodynamic resistance, vehicle mass, road adhesion, and braking force on vehicle motion. The emergency braking logic estimates the required braking distance according to the current vehicle speed and road adhesion conditions, compares it with the real-time obstacle distance and the preset safety margin, and determines whether to activate the braking request. Bidirectional signal transmission between CANoe and Simulink is realized through the CANoe–MATLAB interface. CANoe is used for parameter configuration, signal monitoring, trace recording, graphical display, and data export, while Simulink is responsible for model calculation and function logic execution. Four groups of simulation tests are designed by changing vehicle speed, road adhesion coefficient, vehicle load, and safety distance threshold. Moreover, the braking trigger distance, residual stopping clearance, and logical consistency are selected as the main evaluation indicators.

Experimental results show that the platform can reliably perform scenario configuration, model operation, braking logic triggering, brake actuation, result visualization, and data recording. Under different vehicle speed conditions, as the speed increases from 6 to 12 km/h, the braking trigger distance increases from 1.30 to 1.84 m, whereas the residual stopping clearance decreases from 0.87 to 0.62 m. This indicates that the system can trigger braking earlier when vehicle kinetic energy increases. Under different road adhesion conditions, the braking trigger distance increases and the residual stopping clearance decreases as the adhesion coefficient decreases, showing that the platform can reflect the influence of low-adhesion roads on braking capability. Under different load conditions, as the equivalent vehicle mass increases from 1 830 to 2 200 kg, the braking trigger distance increases from 1.55 to 2.00 m, whereas the residual stopping clearance decreases from 0.62 to 0.38 m, consistent with the increase in vehicle inertia. Under different safety distance thresholds, as the threshold increases from 0.7 to 1.3 m, the braking trigger distance increases from 1.35 to 2.30 m and the residual stopping clearance increases from 0.28 to 0.89 m. These trends are consistent with the basic braking mechanism and the expected logic of low-speed emergency braking.

The CANoe/Simulink cosimulation experimental platform developed in this study effectively supports early-stage verification of low-speed emergency braking functions. The platform integrates parameter configuration, model computation, signal interaction, logic triggering, process visualization, and result recording into a unified experimental environment. Compared with direct real-vehicle testing, it reduces test risk and implementation cost and facilitates the repeated construction of typical low-speed scenarios. The test results under vehicle speed, adhesion coefficient, load, and safety distance conditions demonstrate that the proposed platform can accurately reflect the influence of key variables on braking trigger distance and residual stopping clearance. The platform can be used for preliminary function verification and parameter debugging of low-speed emergency braking systems. Moreover, it can serve as an experimental carrier for courses related to automotive electronics, vehicle control, intelligent vehicle testing, and virtual simulation.