See the View (STV) Project Environmental Perception for Safer and - - PowerPoint PPT Presentation

see the view stv project
SMART_READER_LITE
LIVE PREVIEW

See the View (STV) Project Environmental Perception for Safer and - - PowerPoint PPT Presentation

See the View (STV) Project Environmental Perception for Safer and Reliable Autonomous Driving Arie Gavriely Arie.Gavriely@vayavision.com Co-funded by the Horizon 2020 program of the European Union Proprietary & Confidential | 1


slide-1
SLIDE 1

Proprietary & Confidential | 1 Proprietary & Confidential | 1

See the View (STV) Project

Environmental Perception for Safer and Reliable Autonomous Driving

Arie Gavriely Arie.Gavriely@vayavision.com

Co-funded by the Horizon 2020 program of the European Union

slide-2
SLIDE 2

What we do

Driving Decisions Perception & Fusion Sensors

Vehicle Perception based on raw data fusion with upsampling and AI

Co-funded by the Horizon 2020 program of the European Union

slide-3
SLIDE 3

Proprietary & Confidential | 3

Fusion & Perception

Perception Engine Raw Data Fusion & Detection

⁺ Raw data fusion not objects ⁺ 3D RGBd model

➢Higher detection rates ➢Less false alarms ➢Overcome faults

Camera

. . .

RADAR LiDAR

Co-funded by the Horizon 2020 program of the European Union

slide-4
SLIDE 4

Proprietary & Confidential | 4

Fusion & Detection Framework ⁺ Higher Quality (accuracy and reliability) ⁺ Lower system cost (Sensors, BUS, processing, energy) Sensors Raw Data ⁺ Camera ⁺ Lidar ⁺ Radar Environmental Model

Co-funded by the Horizon 2020 program of the European Union

slide-5
SLIDE 5

Proprietary & Confidential | 5

Detection, Classification, Segmentation

Obstacle detection Road Detection Object BB detection Semantic Segment.

Objects data-base

Objects

Multi-frame Object Tracking

Environmental Model HD RGBd 3D Model

Software Framework

Fusion & Perception

Ego- motion

Up- sampling

Calibrate & Unroll & Match

E-M I/F

Driving Module

Raw data Fusion

Camera

. . .

RADAR LiDAR

Co-funded by the Horizon 2020 program of the European Union

slide-6
SLIDE 6

LIDAR 64 - Before Up-sampling

Co-funded by the Horizon 2020 program of the European Union

slide-7
SLIDE 7

LIDAR 64 - After Up-sampling

3D High Resolution at Low Cost

Co-funded by the Horizon 2020 program of the European Union

slide-8
SLIDE 8

Proprietary & Confidential | 8

Accurate SLAM – Ego-Motion

Co-funded by the Horizon 2020 program of the European Union

slide-9
SLIDE 9

Proprietary & Confidential | 10

Environmental Model

Pedestrian Vehicles

Free Space Free Space Red – Objects

Bird Eye View Detection by Non- DNN Detection by Classification (DNN)

Blue Ground

Co-funded by the Horizon 2020 program of the European Union

slide-10
SLIDE 10

Proprietary & Confidential | 11

Peacocks crossing the street Detecting the unexpected – Non-DNN

slide-11
SLIDE 11

Proprietary & Confidential | 12

Urban Driving in Tel-Aviv

slide-12
SLIDE 12

Proprietary & Confidential | 13

STV Project Objectives

Co-funded by the Horizon 2020 program of the European Union

slide-13
SLIDE 13

Proprietary & Confidential | 14

STV expected and planned performances

Co-funded by the Horizon 2020 program of the European Union

slide-14
SLIDE 14

Proprietary & Confidential | 15

STV Project plan

Co-funded by the Horizon 2020 program of the European Union

slide-15
SLIDE 15

Proprietary & Confidential | 16

STV Schedule and Gantt Chart

Co-funded by the Horizon 2020 program of the European Union

slide-16
SLIDE 16

Proprietary & Confidential | 17 Proprietary & Confidential | 17

Benefits

⁺ Raw Data Fusion

⁺ The optimal method for reliable detection

⁺ Up-Sampling

⁺ Increasing LiDAR/RADAR low resolution with smart algorithms

⁺ Detection and Classification

⁺ Most accurate and reliable detection

Better performance with current sensor set and Roadmap for low cost system

Co-funded by the Horizon 2020 program of the European Union