Autonomous Driving Meetup #5 BERLIN
Details
Autonomous Driving Meetup in in BERLIN hosted by LiagDao! With demos & outstanding talks!
Looking forward to see you there!
Please note that seats are limited. First come first serve :-)
___ Agenda ___
_ LD Data Factory – Intelligent Tool Chain for Training and Testing of ADAS/AD Functions
Dr. Shengguang Lei, Global CTO, LiangDao
_ How to Trust Our Maps? AI-Based Change Detection in HD Maps
Chris Plachetka, AI & Map Expert for Automated Driving, Volkswagen
_ Unlocking L4 Rollout with Teleoperations
Bogdan Djukic, Co-Founder, VP Engineering, Teledrive Experience and Autonomy, Vay
__ Demo __
LiangDao reference system for validation
--> 2 demos: parking solution / camera-LiDAR calibration
--> 2 time slots: 18.00 - 18.30 & 19.45 - 20.15
Registration:
Please register via LinkedIn message to Jannik Dennier* and include your preferred time slot. Participation is confirmed only after receiving a confirmation.
___ Further Details ___
_ If you join the Meetup, you give declaration of consent that pictures will be taken
_ The talks are all held in English and will take ~15-25 minutes each + ~10 minutes Q&A
_ Snacks & drinks will be provided, we'll have socializing breaks
__ Getting There __
Public transport
Bus: M27 (station: Reuchlinstr., distance to venue: 400m); 101/245 (station: Helmholtzstr., distance to venue: 400m)
S-Bahn
S41/S42/S46/S47 (station: Beusselstr., distance to venue: 1,2km); S3/S6/S7/S9 (station: Tiergarten, distance to venue: 1,4km)
Parking
Parking at Spreebogen Plaza is available in the car park, but spaces are limited. We recommend arriving early to secure a spot. Parking outside the building is possible, but please note that there are no spots specifically dedicated to event participants.
__ Session Details __
LD Data Factory – Intelligent Tool Chain for Training and Testing of ADAS/AD Functions
Dr. Shengguang Lei, Global CTO, LiangDao
Abstract: Data is driving the evolution of autonomous driving. Perception system design is a vital step in the development of an autonomous vehicle. For enabling vehicles to perceive the environment, a rapid iteration and optimization of perception algorithm based on extensive scenario data is needed. LD Data Factory is an an end-to-end solution efficiently processes and manufactures data for training and testing of ADAS/AD functions. It features comprehensive data management, a perception software and annotation platform both empowered by AI, as well as automatic data operation and quality assurance. LD Data Factory can reduce 90% of the effort for data labeling compared to existing solutions.
How to Trust Our Maps? AI-Based Change Detection in HD Maps
Chris Plachetka, AI & Map Expert for Automated Driving, Volkswagen
Abstract: Modern automated vehicles (AVs) rely on high-definition (HD) map data to enable automated driving functions. Map-less driving remains an unfulfilled dream at this point, attributed to increased safety when using map data. However, HD maps are prone to errors and become outdated quickly over time, given the ever-changing nature of our surroundings. Ignoring such error modes would be reckless. This talk introduces the need for HD maps even today and AI-based methods for detecting deviations between map data and the real world. Using specialized map evaluation methods can verify map data in adverse conditions, where only degenerated sensor data is available, thereby increasing the safety and availability of the driving function.
Unlocking L4 Rollout with Teleoperations
Bogdan Djukic, Co-Founder, VP Engineering, Teledrive Experience and Autonomy, Vay
Abstract: One of the main reasons for delayed deployment and higher costs of the automated driving is handling the long-tail events. In order to handle edge-cases, AD companies are spending billions and adding years to their roadmaps. This talk is about the concept of "Assisted Autonomy" or how teleoperations can help in commercializing AD products sooner, but still in a safe way.