Vehicle Mode Architecture Case Study

KTH (www.kth.se) – Department of Machine Design

RFI of Vehicle Mode Architecture /RFI (v1.1) – Answers to request for information by KTH Martin Törngren, DeJiu Chen, Lei Feng

6.1 General questions
6.1.1 Does the RFI omit any important aspect of vehicle mode management? If so, please explain.
Answer:

The following aspects are treated in this subsection:

• Consideration of a mode taxonomy

• The definition of the concept of mode itself

• Dependency between vehicle modes and the vehicle variability

• Explicit identification of system stakeholders and their concerns

In our opinion, an important complement to the RFI is the explicit consideration of the taxonomy of various types of modes and the hierarchical organization of
While an architecture can be defined for currently conceived modes, the number of modes will continue to grow with the emerging functions in vehicles. The key to effectively manage vehicle modes is a guideline for defining and organizing them. The guideline works also as an architecture framework for creating customizable mode architectures. To support the evolution of vehicle modes, an effective architecture must allow easy incorporation of new modes and provide pertinent strategies for managing modes of different characteristics. Some relevant aspects for categorizing vehicle modes are listed as follows. The modes corresponding to each aspect are just illustrative examples. A more comprehensive list of vehicle modes is given in the answer of question
In a lockstep execution, a particular vehicle mode should correspond to other vehicle modes or certain states of specific system functions or components. In a synchronous execution, vehicle modes and function/component states are synchronized frequently with each other. In an asynchronous execution, vehicle modes and function/component states work independently and have no demands on state agreement among each other. One effective architectural solution for regulating accesses to shared actuation and computation resources is to introduce arbitration servers. For example, Server- Based Scheduling [8] is a kind of share-driven scheduling scheme that allows the isolation of temporal aspect of applications and thereby promotes dynamic reconfiguration in regards to the management of shared resources (e.g., task and network