Sensor Fusion and ISP Challenges

Security and Active Safety

Mapping and Data Analysis

00 - 00 Local time / 00 - 00 UTC

Raw Sensor Fusion: creating robust environmental model to address functional safety and enable autonomous driving

L2 and L2+ are becoming mainstream and are implemented in many vehicles today. However, the switch to L3 and above is extremely problematic since the responsibility moves from the driver to the AV. This ...Read More

00 - 00 Local time / 00 - 00 UTC

Safety First for Automated Driving (SaFAD)

Through this collective work, we define 12 guiding principles of safe automated driving development for SAE Level 3 and Level 4 automated driving systems, derived from a comprehensive collection of ...Read More

00 - 00 Local time / 00 - 00 UTC

The Hidden Value of Precision Maps

Precision maps are vital to automated driving and a given for robo-taxi companies who need that kind of digital footprint. But do you know that precision maps are useful for making ADAS safer too? ...Read More

00 - 00 Local time / 00 - 00 UTC

In-vehicle Sensor Fusion AI Enabled with Interior Scene Analysis and Modern Automotive-Grade AI Chips

This session covers the latest advancements and advantages of in-cabin sensor fusion AI using modular an calibration-free image, radar and thermal sensor data in real-time. Enabling new real-world use cases for enhanced safety and ...Read More

00 - 00 Local time / 00 - 00 UTC

Towards full autonomy: safety implications for sensor architectures

•Functional Safety: how do L4 requirements differ from ADAS requirements?•Fault tolerant architectures: how can we architect a system that properly detects and reacts to sensor failures? How can we validate such architecture?•SOTIF and scenario-based ...Read More

00 - 00 Local time / 00 - 00 UTC

One reality, many maps: Map applications, deployments, and data sharing

Maps are now a key part of driver assistance systems.  The session will review how maps are being used for ADAS in vehicles, and the data sharing that enables the freshness and accuracy of ...Read More

00 - 00 Local time / 00 - 00 UTC

BMW in-cabin monitoring proof-of-concept with Melexis time-of-flight technology

Sophisticated active safety systems in autonomous vehicles, for example, require awareness of the vehicle interior. The driver’s body and head pose and the number of occupants are among the most useful information that ...Read More

00 - 00 Local time / 00 - 00 UTC

Accelerating ADAS through High Performance, Long Range Configurable LiDAR

Levels 0-2 ADAS has taken off, with features such as automatic emergency braking and blind spot detection proving both effective, and profitable for carmakers and consumers. But getting to L3+ requires an advanced ...Read More

00 - 00 Local time / 00 - 00 UTC

Demo: Implementation of efficient Driver Monitoring Systems

Please join CEVA and partners NextChip and Pathpartner in how to utilize industry leading IP, hardware and software to deliver high performing Driver Monitoring Systems into production. ...Read More

00 - 00 Local time / 00 - 00 UTC

Programming Heterogeneous Systems for Automotive Applications using Quasar

In this talk, we will discuss some key programming principles for heterogeneous systems, in particular how to optimize across components (on the low-level image processing level and on the system level) to answer questions ...Read More

00 - 00 Local time / 00 - 00 UTC

Practical validation of AI within the SOTIF framework

We present an accelerated Bayesian HARA approach leveraging hazardous miles, rendering SOTIF practical.It is possible to integrate FuSa, SOTIF and STPA, into a single common best practice.Whereas FuSa focused on correct implementation against verifiable ...Read More

00 - 00 Local time / 00 - 00 UTC

HD vehicle horizon for automated driving based on the ADASIS standard

Map data enables cars to more accurately localize themselves, ensures a more precise environmental perception and enhances path planning. This allows to extend the sight of the vehicle beyond its sensor horizon, profoundly increasing ...Read More

00 - 00 Local time / 00 - 00 UTC

Are you wondering how to produce a high-precision camera module in the shortest cycle time of less than 15 seconds?

Then join us for a Live Presentation during AutoSens Detroit. We will exclusively present our ProCam TT system in operation and we will show you how the fastest and most precise Active Alignment ...Read More

00 - 00 Local time / 00 - 00 UTC

Improving sensor performance: a look at trends and solutions for optical filters

For most emerging technologies, the performance is expected to be constantly improving while the cost is also expected to be constantly falling.  These same rules apply to the automotive optical sensor market.  A ...Read More

00 - 00 Local time / 00 - 00 UTC

The challenges of architecting a surround-view system for both machine and human vision

This session will give an overview of the challenges faced when architecting a combined driver-assist and surround-view system, where the requirements of both machine vision and human-viewed display need to be met with ...Read More

Coffee Break

00 - 00 Local time / 00 - 00 UTC

ISP adaptation for computer vision

In this presentation we focus on a practical case of using ISP to collect training data. Widely asked question in computer vision society is how to tune ISP for training data collection and how ...Read More

00 - 00 Local time / 00 - 00 UTC

Safety First for Automated Driving: Safe Design Considerations for DNN - Deep Dive

How to ensure functional safety of DNNs is still an open question. This presentation will look at the development steps of DNNs, i.e., Define, Specify, Develop and Evaluate, and Deploy and Monitor from a ...Read More

00 - 00 Local time / 00 - 00 UTC

Perception’s role in creating precautionary driving

Strategies for achieving precautionary drivingImportance of knowing sensor and actuator capabilitiesRisk management in the planning of trajectories ...Read More

00 - 00 Local time / 00 - 00 UTC

Implementation of cost efficient, high performing computer vision systems and how wide-angle systems can deliver an optimised solution

In this discussion, Immervision, CEVA and Pony.ai will offer their viewpoints on how to design the best vision system for cars and autonomous driving. Discussions will include how to choose the right field of view ...Read More

00 - 00 Local time / 00 - 00 UTC

From Perception to a Framework for Verification of Autonomous Behaviour

Perception supplies autonomous systems such as self-driving cars or unmanned aerial vehicles with inputs about the status of the system as well as the surroundings. These inputs are then combined with other information and ...Read More

00 - 00 Local time / 00 - 00 UTC

Commercially Viable Embedded AI for Everyone

r understanding of the status of the embedded AI perception software in relation to the mass market The importance of network size and lean perception network in relation ...Read More

00 - 00 Local time / 00 - 00 UTC

High Dynamic Range Image Sensor Characterization Using a Uniform Source

Data-driven decisions are increasingly reliant on image sensors and systems. Integrating sphere uniform sources provide spectral radiance or irradiance for focal-plane array or complete camera testing. In this presentation we will walk through ...Read More

00 - 00 Local time / 00 - 00 UTC

Latest Solution Demos: Automotive Sensing and Viewing Solutions – Inside and Outside of the Vehicle

OmniVision will demonstrate its latest automotive solutions for both machine vision and viewing applications, including low light performance, RBG-Ir Global Shutter, LFM and HDR, low power consumption, and high resolution: Industry leaders’ latest ...Read More

00 - 00 Local time / 00 - 00 UTC

Perception training data production for multiple sensors via synchronous simulation

rFpro has developed a means to slash the hardware costs associated with large-scale simulation. It ensures that AV and ADAS engineers can perform complex simulations involving multiple sensors. The ground-breaking approach, Data Farming, ...Read More

00 - 00 Local time / 00 - 00 UTC

Photonics-Powered Components Provide Safety, Comfort and Trust in Autonomous Vehicles

This presentation will highlight the latest LED, IR, Laser and VCSEL technologies from OSRAM Opto Semiconductors that enhance the 3rd living space experience by fusing multi-dimensional sensory input to deliver impactful illumination, visualization and ...Read More

00 - 00 Local time / 00 - 00 UTC

Improving Safety with 3D Thermal Ranging

We will discuss the use of thermal imagers which are ideally suited to address these issues and are uniquely able to improve safety. Thermal cameras operate in any lighting condition, are impervious to solar ...Read More

00 - 00 Local time / 00 - 00 UTC

AI assisted sensor data analysis for autonomous vehicles

How Artificial Intelligence can help to reduce the manual efforts in data analysis and validation of autonomous Vehicle systems and sensorsWhat are the different deep learning and machine learning techniques for such data analysisDifferent ...Read More

Panel Discussions

00 00 Local time / 00 00 UTC

Panel discussion: Cybersecurity in automotive

Join this all-women panel to explore how to approach cybersecurity in your day-to-day product development and how engineering and cybersecurity should go hand-in-hand. Cybersecurity by design will also be discussed, including ways to create and …Read More

00 00 Local time / 00 00 UTC

Panel discussion: Do we need thermal cameras in the sensor mix for autonomous driving?

There is much discussion around which sensor combinations will realise autonomous driving. Can cameras do it alone? Is LiDAR a necessity? Do thermal cameras fill the AV sensor gap? …Read More

00 00 Local time / 00 00 UTC

Panel discussion: Challenges of shrinking AI down to in-car units

discussion of the requirements for hardened, highly reliable code, resource constrained environments and not-always-on network connectivity and how these impose a new set of requirements on both, the developers of AI Frameworks as …Read More