Introduction to Automated Vehicle Safety: Multi-Agent, Functional Safety, and SOTIF C1950
Fatal accidents involving automated vehicles have made it clear that safety is paramount to their acceptance, testing, verification, validation, and deployment. In fact, safety has been ranked as the number one concern for the acceptance and adoption of automated vehicles, and understandably so, since safety has some of the most complex requirements in the development of such vehicles. However, there are many misconceptions involving safety and the concept of safety as applied to automated vehicles.
This two-day course will help characterize the nature of safety and the fundamental technology needed by most people involved in the design, development, testing, operation, and deployment of automated vehicles. The course will enable participants to envision a future where the safety of automated vehicles is well understood and will guide all stakeholders in the development and use of safe automated vehicles. You’ll learn the main attributes of safety as applied to automated vehicles, including the three types of safety: Functional Safety, Safety of the Intended Functionality (SOTIF), and Multi-agent safety. The discussion will enable participants to conceive of the various applicable design aspects of safety, clarify the role of SOTIF and multi-agent safety, and address the development of multi-agent safety using a probabilistic and stochastic framework.
Participants should bring a laptop computer for five, in-class exercises designed to ensure application and retention. Learners will complete the exercises during class and will get assistance and feedback from the instructor.
Learning Objectives
By attending this seminar, you’ll be able to: - Articulate the concepts of hazard, risk, risk assessment, and risk reduction
- List and describe the most fundamental ideas and techniques of Functional Safety, SOTIF, and Multi-agent safety
- Describe the salient features and application of the ISO 26262 standard
- Develop a Functional Safety Concept for a specific subsystem of an AV
- Summarize the main NHTSA safety guidelines for autonomous vehicles
- Summarize the main elements to be addressed while writing a safety report for an autonomous vehicle
Who Should Attend
This course will be especially valuable for those needing to address safety issues in the design of ADAS (advanced driver assistance systems) and automated vehicles. Participants should have a mechanical, electrical, or computer engineering or computer science degree.
Exclusive eBook Offer: Your course participation qualifies you for a specially-priced bundle of five books that explore automated vehicle safety concepts and technologies, authored by course instructor, Dr. Juan Pimentel. Click here for a summary of the series and links to a description of each volume. Details on this exclusive offer are sent to learners upon completion of the respective course offering.
Module I
- Safe Autonomous Vehicle Platform: Safety Critical Components
- Control functionality
- Perception system
- Computing platform
- Risk Classification (Automotive Safety Integrity Level: ASIL)
- Preliminary hazard analysis (PHA)
- Safety Functions, Safety Goals
- Overview of Functional Safety
- Functionality
- Perception system
- Computing platform
- AV platform
The assignment consists in developing an autonomous vehicle top level design and performing a PHA (preliminary hazard analysis). The design should specify:
- Overview of ISO 26262
- Development of the Functional Safety Concept
- Failure identification: random and systematic failures
- AV safety critical subsystems
- List of errors, faults, failures, hazards
- AV safety goals
- Assignment of ASILs
- Risk mitigation and risk reduction
- Discussion of Assignment #1
Assignment #2: Description and Requirements
The assignment consists in developing and documenting a Functional Safety Concept (FSC) for the design of assignment #1
- Challenges in the Application of ISO 26262 for Autonomous Vehicles
- Perception system issues (SOTIF)
- End-to-end considerations
- Computing and communication issues
- Behavioral safety considerations
- Introduction to Behavioral Safety
- Introduction to SOTIF
- Discussion of Assignment #2
The assignment consists in evaluating the safety hazards that happens when a vehicle shares the road with other vehicles
DAY TWO
Module IV
- Behavioral Safety
- Sharing the road with others
- Vehicle Dynamics Considerations
- Accidents: Fault, Blame, Guilt
- Responsibility Sensitive Safety (RSS)
- Ego vehicle
- Safe actions/behaviors
- Absolute safety
- Safety Guarantees, Cautious Driving
- Discussion of Assignment # 3
The assignment consists in specifying the main safety critical functions (SFC) of your autonomous vehicle design and for each safety critical function, design a sub-system for risk reduction. For each SCF, specify its ASIL value and include the following: Functional Safety; Safety of the Intended Functionality (SOTIF); Behavioral (multi-agent) Safety
Module V
- Guaranteeing Multi-agent Safety
- Safe longitudinal distance
- Safe cut-in of the ego vehicle
- Safety of the Intended Functionality (SOTIF)
- Role of Governments in AV safety
- NHTSA Safety Guidelines
- Discussion of Assignment #4
The assignment consists in specifying the main safety features and processes of an autonomous vehicle and develop a corresponding safety report according to NHTSA guidelines to include: Operational Design Domain (ODD), Object and Event Detection and Response (OEDR), Fallback (Minimum Risk Condition)
Module VI
- System Safety
- Designing a capable system
- Analyzing safety performance
- Applying development processes
- Writing a Safety Deport
- Operational design domain (ODD)
- Object and event detection and response (OEDR)
- Fallback (Minimum Risk Condition)
- Discussion of Assignment # 5
- Summary/Take Away
Juan R. Pimentel
Dr. Juan R. Pimentel is a Professor of Computer Engineering at Kettering University in Flint, Michigan. He is an expert in the Internet of Things (IoT), Industrial Internet, systems engineering, safety-critical systems, automated vehicles, and the safety of autonomous vehicles and is a recognized international expert in the areas of industrial communications, real-time and dependable systems, and autonomous vehicle safety. He has written books on industrial networking, multimedia systems, and safety-critical automotive systems. Dr. Pimentel has also performed extensive international consulting and conducted professional training courses in North and South America, Europe, Asia, and the Middle East. He is an expert witness on patent infringement cases involving automotive systems, industrial communications, and IoT.
Dr. Pimentel has performed research at institutions around the world such as the Franuhofer Institute, Germany; INRIA, France; University of Padova, Italy; Universidad Polictecnica de Madrid and Universidad Carlos III de Madrid, Spain; Universidad de los Andes, Colombia; and UTEC, Peru. In 2007 he received the “Distinguished Researcher Award” from Kettering University for contributions in the area of industrial communication systems and automotive systems. He has written over 86 peer reviewed papers at international conferences and Journals, primarily the IEEE and SAE.
As a 1980 graduate of the University of Virginia, additional accomplishments include the co-development of the application layer for Profibus (with Siemens), and the development of FlexCAN, a CAN-based dependable architecture for safety-critical applications. In the last few years he has been involved with various projects dealing with automated vehicles including design, simulation, testing, functional safety, and developing online training materials. He is a faculty advisor to the Kettering University team participating in the AutoDrive autonomous vehicle competition organized by SAE International and General Motors. One of his latest projects involve the development of techniques and methodologies to design automated vehicles with a sufficient level of safety.
Fees: $1415.00
SAE Members: $1132.00 - $1274.00
1.3 CEUs
You must complete all course contact hours and successfully pass the learning assessment to obtain CEUs.
If paying by a credit card, click the Register button above. If paying by any other method or for general inquiries, please contact SAE Customer Service 1-877-606-7323 (724-776-4970 outside the U.S. and Canada) or at CustomerService@sae.org.
