Understanding Vehicle E/E Architecture Topologies for Automated Driving: System Partitioning and Tradeoff Parameters

Mody, Mihir; Jones, J. P.; Chitnis, Kedar; Sagar, Rajat; Shurtz, Gregory; Dutt, Yashwant; Koul, Manoj; Biju, MG; Dubey, Aish

doi:10.2352/issn.2470-1173.2018.17.avm-358

Cited by 14 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) Domain Centralised E/E Architecture -the "in development" architecture: The increasing number of ECUs and the attention to software qualities as scalability, robustness, and maintainability led to a new and more structured architecture [11]. The Domain Centralised E/E architecture adopts a layered architectural style and introduces the concept of domain [17], which is used for grouping ECUs, e.g. Powertrain for the control of engine and batteries, Infotainment for the control of displays, entertainment system, and navigation, etc.…”

Section: B Electrical and Software Architecture Of Automotive Systemsmentioning

confidence: 99%

Agile Beyond Teams and Feedback Beyond Software in Automotive Systems

Ågren,

Heldal,

Knauss

et al. 2022

Preprint

View full text Add to dashboard Cite

In order to increase the ability to build complex, software-intensive systems, as well as to decrease time-to-market for new functionality, automotive companies aim to scale agile methods beyond individual teams. This is challenging, given the specifics of automotive systems that are often safety-critical and consist of software, hardware, and mechanical components. This paper investigates the concrete reasons for scaling agility beyond teams, the strategies that support such scaling, and foreseeable implications that such a drastic organizational change will entail. The investigation is based on a qualitative case study, with data from 20 semi-structured interviews with managers and technical experts at two automotive companies. At the core of our findings are observations about establishing an agile vehicle-level feedback loop beyond individual teams. (I) We find that automotive OEMs aim to decrease lead-time of development. (II) We also identify 7 strategies that aim to enable scaled-agile beyond teams. (III) Finally, we extract 6 foreseeable implications and side-effects of scaling agile beyond teams in automotive. By charting the landscape of expected benefits, strategies, and implications of scaling agile beyond teams in automotive, we enable further research and process improvements.Managerial Relevance Statement-From interviews with highand mid-level managers in two automotive OEMs, we derive qualitative insights on scaling agile ways-of-working beyond single teams. We analyze what abilities automotive OEMs are seeking to achieve from applying agile at this scope, strategies proposed for scaling, and implications anticipated of the strategies. We believe that the findings of this study can provide guidance, on the trade-offs between different approaches when scaling agile, to managers in automotive or other domains with a mix of software, hardware, and mechatronics. I. INTRODUCTIONAgile ways-of-working initially focused on small teams developing software. The success of agile approaches in this context has, however, led to adoption also in the development

show abstract

Section: B Electrical and Software Architecture Of Automotive Systemsmentioning

confidence: 99%

Agile Beyond Teams and Feedback Beyond Software in Automotive Systems

Ågren,

Heldal,

Knauss

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The Automated driving (AD) systems implement highway driving and value parking functions at multiple automation levels (L2-L5) [1][2]. The figure 1 shows the functional block diagram to achieve these functions.…”

Section: Introductionmentioning

confidence: 99%

Open source deep learning inference libraries for autonomous driving systems

Desappan¹,

Anand²,

Pramod³

et al. 2022

Self Cite

View full text Add to dashboard Cite

Automated driving functions, like highway driving and parking assist, are increasingly getting deployed in high-end cars with the goal of realizing self-driving car using Deep learning (DL) techniques like convolution neural network (CNN), Transformers etc. Traditionally custom software provided by silicon vendors are used to deploy these DL algorithms on devices. This custom software is optimal for given hardware, but supports limited features, resulting in-flexible for evaluating various deep learning model architectures tradeoffs by means of rapid prototyping. This paper proposes usage of various open-source deep learning inference frameworks to quickly deploy any model architecture without any performance/latency impact. The proposed solution consists of automatic Graph Partitioning, Post-Training Quantization and Optimal Tensor Management. The proposed solution has been implemented with three open-source inference frameworks namely Tensorflow-Lite, TVM/Neo-AI-DLR and ONNX Runtime running on Linux OS. The proposed solution using opensource frameworks provides ease of use and improved coverage for network types due to fall back for unsupported features from custom software provided by silicon venders.

show abstract

“…The warnings can also be used to activate pre-collision safety measures, such as occupant safety belt pre-tensioners, airbag arming, and increased seat side and leg bolstering [33,34]. This EOAM framework should also embrace operational design domain (ODD) and dynamic driving task (DDT)-based architecture [28][29][30][31] that allows for specific domains to take care of unique tasks that occur less frequently than normal driving conditions; one of the domains could manage emergency maneuvers involving large-magnitude lateral dynamics with low processing and execution times.…”

mentioning

confidence: 99%

“…These buffers are used not only to avoid any contact with the ARO during the maneuver but to also reduce the occupant fear and increase their visual comfort in the situation during the maneuver. The unique DMM phase diagram in this paper embraces the idea that this part of the software logic would be integrated into the system-level AV architecture as a design domain [28][29][30]. This DMM would be a sub-module of the middle-level EOAM framework module.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Autonomous Vehicle Emergency Obstacle Avoidance Maneuver Framework at Highway Speeds

Lowe,

Guvenc

2023

Electronics

View full text Add to dashboard Cite

An autonomous vehicle (AV) uses high-level decision making and lower-level actuator controls, such as throttle (acceleration), braking (deceleration), and steering (change in lateral direction) to navigate through various types of road networks. Path planning and path following for highway driving are currently available in series-produced highly automated vehicles. In addition to these, emergency collision avoidance decision making and maneuvering are another key and essential feature that is needed in a series production AV at highway driving speeds. For reliability, low cost, and fast computation, such an emergency obstacle avoidance maneuvering system should use well-established conventional methods as opposed to data-driven neural networks or reinforcement learning methods, which are currently not suitable for use in highway AV driving. This paper presents a novel Emergency Obstacle Avoidance Maneuver (EOAM) methodology for AVs traveling at higher speeds and lower road surface friction, involving time-critical maneuver determination and control. The proposed EOAM framework offers usage of the AV’s sensing, perception, control, and actuation system abilities as one cohesive system to avoid an on-road obstacle, based first on performance feasibility and second on passenger comfort, and it is designed to be well integrated within an AV’s high-level control and decision-making system. To demonstrate the efficacy of the proposed method, co-simulation including the AV’s EOAM logic in Simulink and a vehicle model in CarSim is conducted with speeds ranging from 55 to 165 km/h and on road surfaces with friction ranging from 1.0 to 0.1. The results are analyzed and interpreted in the context of an entire AV system, with implications for future work.

show abstract

Understanding Vehicle E/E Architecture Topologies for Automated Driving: System Partitioning and Tradeoff Parameters

Cited by 14 publications

References 0 publications

Agile Beyond Teams and Feedback Beyond Software in Automotive Systems

Agile Beyond Teams and Feedback Beyond Software in Automotive Systems

Open source deep learning inference libraries for autonomous driving systems

Autonomous Vehicle Emergency Obstacle Avoidance Maneuver Framework at Highway Speeds

Contact Info

Product

Resources

About