Embedded software designs, such as those used in avionics and automotive systems, have become so complex to develop and create that a design environment without coordination is becoming problematic for all developers involved.
Model-based design (MBD) is a whole new approach currently being applied in the design of embedded software. It has become de rigueur in highly complex design applications such as guidance systems, engine controls, autopilots, anti-lock braking systems, to name a few. A new eInfochips’ report by Mangesh Kale and Anupama Shendage, “Model-Based Design for Embedded Software,” details the processes and advantages of MBD.
MBD has evolved to overcome various difficulties and complexities that typically arise during the design lifecycle of embedded software for closed-loop control systems. Such software needs to be designed in an iterative manner with extensive involvement of multi-disciplinary teams. In most practical scenarios, the need for embedded software design and testing has to start early – before physical prototypes and systems are made available. Using traditional design processes, the discovery of design and requirements errors found late in the design cycle can lead to expensive delays.
The MBD framework aims to address these issues early on in the design phase while significantly minimizing the rework involved in later phases of lifecycle. In traditional design processes, the design information is usually communicated and managed in the form of text-based documentation. Frequently, such documentation is difficult to understand and subject to interpretation bias. The embedded code is created manually from specifications and requirements documents, hence, leading to a time-consuming and error-prone process. There is also little tracking to ensure that changes are implemented correctly.
According to our research, MBD, when used effectively, is able to provide a single design environment so that developers can use a single model of their entire lifecycle for data analysis, model visualization, testing and validation, and ultimately product deployment, with or without automatic code generation. This enables product companies to rapidly roll out their innovations and new prototypes, and with fewer design iterations.
An Improved, Cost-Effective Approach
The strategic advantage of using MBD lays in techniques like simulations and small scale models that can greatly reduce the costs and risks of production. This is especially true in the high-tech manufacturing industry where embedded systems seriously impact mechanical, electrical, pneumatic and hydraulic machines. However, MBD can also be used effectively and economically for less complex designs. The approach creates a structure for software reuse that allows established designs to be effectively and reliably upgraded in a more cost effective manner.
Engineering firms often have a realistic grasp of the hard work that a typical embedded project (even in larger firms) entails. Smaller, productive teams with tight deadlines often scope these projects. However, this approach results in a terrible work-life balance for all engineers concerned and leads to higher costs from additional system rework. One way around this problem is to alter the workflow in such a way that there is a need for minimal coding. Using intelligent algorithms and behavioral models, it is possible to predict how the overall system would perform in actual conditions – temperature, pressure etc. Thus, eliminating unnecessary costs.
Using traditional design processes, the discovery of design and requirements errors found late in the design cycle can lead to expensive delays. The MBD framework aims to address these issues early on in the design phase while significantly minimizing the rework involved in later phases of lifecycle. MBD also creates a cost-effective structure comprising models and simulations, for software reuse that allows established designs to be reliably upgraded. The benefits of MBD add up to a significant advantage for embedded systems developers.
At eInfochips, Sayak Boral is Assistant Marketing Manager, Mangesh Kale is Senior Solution Architect and Key Accounts Manager, and Anupama Shendage is an engineer. Click here to access the new report, “Model-Based Design for Embedded Software.”