Scientific articles on Safety-driven Behaviour Management

Hazard recognition and risk assessment in open and non-predictive environments are essential for autonomous and semi-autonomous systems for proper decision making and action selection. Neither existing safety standards nor situation modeling is commonly considered in that context. A novel approach denoted as a Safety-Driven Behavior Management (SDBM) combines the safety standards-oriented hazards analysis and the risk assessment approach with the machine learning-based situation recognition. It can help to overcome existing autonomous systems-related safety issues and support the safety argumentation. This article summarizes the concept, possible applications, and first test results, which are introduced detailed [1] and [2].

What is Safety-Driven Behavior Management?

The goal of the Safety-Driven Behavior Management (SDBM) system [1, 2] is to manage and control the behavior of an autonomous or semi-autonomous system during its interaction with the environment. The SDBM consists of several modules, which can be re-used in different contexts. It allows the usage of the SDBM for higher automation levels [3] but also the design of assistance systems. The major tasks solved in different modules are:

• acquisition of data and extraction of information concerning environmental objects and technical ego-system state,

• structuring and assessing the information representing a situation,

• identification of hazards and risk assessment,

• situation recognition verifying and improving risk assessment,

• planning, reasoning and finding an optimal action considering the mission goal as well as safety.

Read full scientific articles here

[1] G. Hägele and A. Sarkheyli-Hägele, “Situational hazard recognition and risk assessment within safety-driven behavior management in the context of automated driving,” in 2020 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA). IEEE, 2020 (The article is available to IEEE members or can be purchased on that webpage.)

Available here

[2] G. Hägele and A. Sarkheyli-Hägele, “Situational risk assessment within safety-driven behavior management in the context of UAS,” in The 2020 International Conference on Unmanned Aircraft Systems (ICUAS). IEEE, 2020. (The article is available to IEEE members or can be purchased on that webpage.)

Available here

[3] G. Hägele. “Contribution to realization and test of a fall-back layer for safe, autonomous, and actionflexible systems.” Ph.D. thesis, Duisburg-Essen Publications online, 2018

Available here

Benefits of Safety-driven Behaviour Management for the Semcon customer

Especially Semcon’s customers can benefit from the novel approach. The modularity of the Safety-Driven Behavior Management (SDBM) allows its usage as a construction kit for safety-critical applications. For Semcon’s customers, it means reduced product design and development time, leading to the reduction of project costs.

The ability for the situational recognition of hazards, assessment of risks, and maintaining the acceptable risk level during the interaction with the environment supports significantly the argumentation that the autonomous system the SDBM is applied for is safe. This fact gives Semcon’s customers a clear advantage for the compilation of the product’s proof of safety like the safety case and its certification.

With the this approach, Semcon can support current and future customers on their exciting journey towards safe autonomous systems.