CarSim与ADAS集成：参考手册详解

需积分: 0 173 浏览量更新于2024-06-22 1 收藏 18.23MB PDF 举报

"CarSim+ADAS+参考手册是一份由Mechanical Simulation出版的技术文档，主要涉及CarSim软件与高级驾驶辅助系统(ADAS)的集成应用。这份手册旨在帮助用户理解和利用CarSim中的ADAS功能，包括传感器、目标物体、参考路径、变量和参数设置等关键元素。" CarSim是一款广泛应用的车辆动力学仿真软件，它能够模拟各种车辆的动态行为。在集成ADAS系统后，CarSim能够模拟高级驾驶辅助功能，如碰撞预警、自适应巡航控制和车道保持等。这份手册详细介绍了如何在CarSim环境中设置和操作ADAS传感器，以及处理移动和固定对象。 ADAS传感器和目标对象是ADAS功能的核心部分。手册涵盖了不同类型的传感器，如雷达、激光雷达(LiDAR)和摄像头，这些传感器用于检测车辆周围的环境，识别行人、自行车手、动物以及其他车辆。此外，手册还讨论了目标物体的定义，包括它们的位置、运动状态以及如何在仿真环境中设定。参考路径是ADAS系统进行导航和决策的基础。用户可以设置车辆和对象沿特定路径移动，以模拟实际道路情况。这部分内容详细讲解了如何创建和管理参考路径，以便进行复杂的交通场景仿真。手册的变量和参数部分强调了如何配置和调整对象的属性，如速度、加速度、位置等，以满足特定的仿真需求。同时，它也介绍了如何使用“Multiple Moving Objects Screen”来管理多个独立运动的对象，以及如何通过“Single Moving Object (Custom) Screen”定制单个移动对象的行为。对于复杂形状，如多边形形状坐标，手册提供了指导，让用户能够精确地定义物体的几何形状，以更真实地模拟道路障碍物或非规则形状的目标。动画对象如行人、自行车手和动物的模拟是ADAS仿真中的重要方面。这部分内容详细描述了如何使这些对象在模拟中动态表现，以增加场景的真实感。 ADAS范围和跟踪传感器的设置是理解传感器性能的关键。手册阐述了检测几何学和遮挡的概念，以及如何设置传感器对距离、方位角和俯仰角的敏感度。同时，它还解释了如何使用“Sensor Reflectiveness Map Screen”来调整传感器的反射率特性，以反映不同材料对传感器信号的影响。传感器的输入输出变量是连接仿真结果与外部系统的桥梁。手册详述了如何导入和导出这些变量，以便与其他软件系统集成，例如进行数据分析或控制系统设计。计算速度部分可能涉及如何优化仿真过程，以在不影响精度的情况下提高计算效率，这对于处理大量数据和复杂场景至关重要。 "CarSim+ADAS+参考手册"为用户提供了全面的指南，帮助他们充分利用CarSim的ADAS功能进行驾驶安全研究、车辆性能评估和自动驾驶系统开发。这份手册对于汽车工程、交通研究和ADAS系统开发的专业人士来说，是一份宝贵的参考资料。

ADAS Sensors and Moving Objects

Table of Contents (All Screens) 10 / 59 VehicleSim Browser Reference Manual

Table 2. Parameters that determine how an object location is controlled.

Parameter Description Parameters

enabled

ROAD_ID_OBJ

ID of road surface used to

locate object vertically

OPT_ROAD_PITCH,

OPT_ROAD_ROLL

OPT_ROAD_PITCH

Booleans: set pitch or roll to

be parallel to the road

surface?

OPT_ROAD_ROLL

PATH_ID_OBJ

ID of path used to locate

object horizontally

LTARG_ID_OBJ,

OPT_SPEED_OBJ

LTARG_ID_OBJ

ID of LTARG dataset to locate

object laterally relative to

path

The parameter ROAD_ID_OBJ provides the ID for a road surface, and determines how the object

is located vertically. If ROAD_ID_OBJ is set to zero, then Z_Obj_o is used to control height, and

the variables PitchOo and RollO_o (the object pitch and roll angles) may be set as needed.

Otherwise, ROAD_ID_OBJ should be the ID number of a road dataset that will be used to locate

the object vertically.

When ROAD_ID_OBJ is given a valid road surface ID, there are two more parameters that

determine how the object is oriented in pitch and roll: OPT_ROAD_PITCH and

OPT_ROAD_ROLL. Both are Booleans. When set to 1, the pitch or roll angle is set to match the

road surface at the location of the object control point. For example, this would be the expected

orientation of an object used to represent a traffic vehicle. When set to 0, the angles are not set

automatically. For example, pedestrians tend to stand upright, even if the road surface is not level.

If when ROAD_ID_OBJ is set to zero, then a road surface is not used; in this case, the parameters

OPT_ROAD_PITCH and OPT_ROAD_ROLL are not relevant and do not appear in the Echo file.

The parameter PATH_ID_OBJ determines how the object is located horizontally. If

PATH_ID_OBJ is set to zero, then the variables X_Obj_o and Y_Obj_o are used to locate the

object, and the variable YawO_o is used to control the yaw. (The variables can be set directly as

with parameters, or set with VS Commands, or imported from external software.) Otherwise,

PATH_ID_OBJ should be the ID number of a path dataset that will be used to locate the object

with S-L coordinates associated with the identified path.

When PATH_ID_OBJ identifies a valid path ID, then the parameter LTARG_ID_OBJ may be

used to specify an LTARG dataset for locating the object laterally relative to the path. Set

LTARG_ID_OBJ to zero if there is no lateral offset. If PATH_ID_OBJ is set to zero, then

LTARG_ID_OBJ is ignored and is not shown in the Echo file.

In addition to the parameters just described, the parameter OPT_RECYCLE_OBJ enables/disables

recycling of objects that go out of view. Recycling is supported fully by the two Moving Objects

screens, and is described along with the GUI controls (see the section The Multiple Moving

Objects Screen, page 14). The parameter OPT_SHAPE_OBJ determines what kind of shape the

object has with respect to acting as a target for an ADAS sensor. These options are described in the

section The Single Moving Object (Custom) Screen (page 31).

ADAS Sensors and Moving Objects

Table of Contents (All Screens) 11 / 59 VehicleSim Browser Reference Manual

Control parameters for speed and offtracking

OPT_SPEED_OBJ is VS Command (with an associated parameter) used to determine the role of

the speed of an object when the object is linked to a path via the PATH_ID_OBJ parameter. When

PATH_ID_OBJ specifies a path, the command/parameter OPT_SPEED_OBJ is shown in the Echo

file, possibly with the other parameters listed in Table 3. If PATH_ID_OBJ is set to zero, then

OPT_SPEED_OBJ has no effect, and it is not shown in the Echo file.

Table 3. Parameters that determine how speed is used.

Parameter

Description

Parameters

enabled

OPT_SPEED_OBJ

Command/parameter that

specifies how V_Obj_

is used

L_WB_OBJ,

SPEED_ID_OBJ,

OPT_ACCEL_OBJ

L_WB_OBJ

Wheelbase used for offtracking

SPEED_ID_OBJ

ID of SPEED_TARGET dataset; if

ID < 0, use negative speed.

If either PATH_ID_OBJ or OPT_SPEED_OBJ is set to zero, then V_Obj_o is calculated using

the differences in the current and previous time step for values of X_Obj_o and Y_Obj_o.

However, if both PATH_ID_OBJ and OPT_SPEED_OBJ have non-zero values, then the station

along the path is installed as a state variable (ODE SV) whose derivative is calculated by

numerically integrating V_Obj_o.

The OPT_SPEED_OBJ Command has three valid settings (0,1,2) with the following effects:

0. Station S_Obj_o is constant or set directly with a VS Command or by importing

Imp_S_Obj_o from external software. No built-in ODEs are used for the object.

1. An ODE is added to the model in which the derivative of S_Obj_o is V_Obj_o.

V_Obj_o may be a constant, it may be calculated automatically with the built-in function

SPEED_TARGET as described below, it may be updated each time step with a VS

Command, it may be imported from external software as Imp_V_Obj_o, or another ODE

may be added to the model in which the derivative of V_Obj_o is A_Obj_o (as described

in the next subsection).

2. All options for option 1 may be used. In addition, an ODE the derivative of yaw angle

YawO_o is calculated to simulate offtracking.

If OPT_SPEED_OBJ is set to a value of 1 or 2 and OPT_ACCEL_OBJ (described later) is zero,

then V_Obj_o should be set directly. It may be set with a VS Command, or by import from external

software, or by using the SPEED_TARGET Configurable Function whose dataset is specified with

the parameter SPEED_ID_OBJ. If SPEED_ID_OBJ is not zero, then the sign of the ID is used to

apply the built-in function SPEED_TARGET as follows:

V_Obj_o = SPEED_TARGET(S_Obj_o, T, I) { SPEED_ID_OBJ > 0 }

V_Obj_o = –SPEED_TARGET(S_Obj_o, T, I) { SPEED_ID_OBJ < 0 }

ADAS Sensors and Moving Objects

Table of Contents (All Screens) 12 / 59 VehicleSim Browser Reference Manual

where I is the SPEED_TARGET dataset whose SPEED_TARGET_ID parameter matches the

absolute value of SPEED_ID_OBJ. For example, if SPEED_ID_OBJ = -1234, and

SPEED_TARGET_ID(7) = 1234, then the SPEED_TARGET function would be evaluated each

time step with I = 7, and V_Obj_o would be set to the negative of that value.

When OPT_SPEED_OBJ is set to 2 to enable offtracking, the derivative of yaw angle YawO_o is

calculated internally with the ODE:

d(YawO_o)/dt = –V_Obj_o • sin(HeadO_o) /L_WB_OBJ

where L_WB_OBJ is a wheelbase parameter. This is a 2-point model (bicycle) with no slipping.

In this mode of operation, heading is calculated automatically as the difference between the object

yaw and the path yaw:

HeadO_o = YawO_o – PathYaw

where PathYaw is the path yaw with a possible contribution from a linked LTARG dataset, as

calculated with the VS Command PATH_YAW_ID.

Figure 10 shows an example made in TruckSim in which a moving object (the truck with a black

cab) goes through an intersection just in front of the simulated vehicle (orange cab and yellow force

arrows). Notice that the right-rear wheel of the leading truck (the moving object) is at the curb, due

to offtracking, while the right-front wheel is following the specified target path closely. In this

example, the object CP is at the center of the front axle, and L_WB_OBJ is the distance between

the two axles.

Figure 10. Moving object and ego vehicle both showing offtracking behavior for a 90° turn.

The X-Y plots compare the tracking of the four points of the moving object with the four wheels

of the full simulated vehicle, showing that they are qualitatively similar. They do not match exactly,

mainly because the control point in the moving object follows the specified path perfectly, while

the vehicle uses a preview path follower with some lag. (The plots show a lateral displacement for

the front vehicle of about 0.4m of the center of the front axle relative to the target path.) The full

simulation also includes secondary effects such as scrubbing of tire, kinematical steering, etc.

剩余58页未读，继续阅读

不爱思考的汽车

粉丝: 1
资源: 6

CarSim与ADAS集成：参考手册详解

汽车紧急避障仿真系统：Carsim+Simulink实践案例

CarSim与Simulink：ADAS智能驾驶环境搭建与控制策略详解

"CarSim制动系统参考手册：控制、压力、模型详解

Carsim+制动系统+参考手册

Carsim+转向系统+参考手册（2019）

CarSim+空气动力学+参考手册

CarSim+悬架系统+参考手册（2019）

carsim+simulink联合仿真实现变道 包含路径规划 carsim+simulink联合仿真实现变道 包含路径规划算法+mpc轨迹跟踪算法 可选simulink版本和c++版本算法 可以适用于

Carsim+Prescan+Simulink联合仿真入门

Carsim+Prescan+simulink_联合仿真.docx

最新资源

carsim+simulink联合仿真实现变道包含路径规划 carsim+simulink联合仿真实现变道包含路径规划算法+mpc轨迹跟踪算法可选simulink版本和c++版本算法可以适用于