Toronto Pearson will be the first airport in the world to trial the HEXWAVE walkthrough security detection portal. Several key members of the Liberty Defense team, including CEO Bill Frain, have previously been involved in the development and mass deployment of key screening technologies that are commonly found today at airports around the world. Bill Frain said:

“The HEXWAVE is planned to act as an additional layer of security at airports, extending the perimeter of the screened area to include the entrance to the airport itself. As airports continue to look at innovative ways to improve security, technology like HEXWAVE provides the ultimate solution.”

This strategy is known as a layered defense strategy, and it makes it possible for security teams to detect threats at the perimeter of a property without obstructing the movement of large groups of people.

HEXWAVE uses 3D radar imaging and artificial intelligence, primarily to detect and identify concealed weapons, as well as other threats and anomalies, whether they are metallic or nonmetallic.

AutoSens Awards originate from 2015. Since then, they have been selecting the most innovative companies and products leading the global autonomous driving industry. This year, StradVision won the Gold Award in ‘Best in Class Software for Perception Systems‘ for its self-developed deep learning-based camera perception software SVNet, at the virtual awards ceremony held on November 24th. Junhwan Kim, CEO of StradVision, commented:

“I am very excited to win one of the most prestigious awards in the autonomous driving industry. StradVision is striving to provide camera perception solutions that achieve the best performance and the highest efficiency. We’ll keep moving forward to spur technology development and business expansion, so the benefits of ADAS and autonomous driving solutions will be accessible to all drivers and pedestrians.”

SVNet is а state-of-the-art perception software that supports ADAS features of driving automation levels 2 and 3 and 4. The software delivers unmatched high performance and efficiency that enables cutting-edge deep learning-based camera perception with minimal computation and power consumption.

What is more, SVNet provides over 30 features for recognition of road environments such as other vehicles, pedestrians, traffic lights and signs, and animal detection. It offers high compatibility for 14 different SoC platforms and enables surround vision systems by integration with multiple sensors such as LiDAR and RADAR.

StradVision was established in 2014, and has been leading the development of autonomous driving technology via its AI-based vision perception software. The SVNet software is currently being deployed in over 50 vehicle models from 13 OEMs worldwide.

Photo Credits: StradVision

Autonomous vehicles rely on three main technologies, the first of which is high-precision localization, robust and reliable environment perception and last but not least the prediction and motion planning algorithms for safe manoeuvres. The motion planner is easier to achieve as long as the accuracy of the localization and perception data is higher.

Photo Credits: Qualcomm

• The unified localization solution

An autonomous vehicle must be able to localize itself within a few centimetres of accuracy to plan and execute complex manoeuvres required in real-world road conditions. A way to reach higher levels of accuracy is by HD Maps. Qualcomm shared:

“Our approach is to develop advanced and optimized modules that solve such challenges for our automotive OEM customers and partners.”

One such component is Qualcomm Vision Enhanced Precise Positioning (VEPP) software, which can provide lane-level positioning and a precise heading in the global coordinates by tightly fusing GNSS, IMU, camera, and vehicle odometry information in real-time and in virtually all environments.

The localization solution fuses the global pose with camera perception and HD map information and also provides an ability to generate a local map, by producing identification of gaps or errors in HD maps on the fly — all while the vehicle can operate in an autonomous mode.

• Novel techniques for efficient environment perception

One of the main focus areas for Qualcomm is to scale environmental perception technologies in a more efficient way.  Their aim is to harness all sensor types including cameras, radar, and lidar, as well as maps to build novel ways of solving environment perception challenges.

There is multisensor data fusion, which includes data association, as well as object-state estimation. Data association is basically the ability of the autonomous vehicle to correlate data from different views. State estimation involves tracking real-world objects and estimates their size, speed, and position relative to the autonomous vehicle itself.

Another Qualcomm’s focus area is the accurate prediction of the pose and tracking the size of objects on the road with 3D estimations. This happens by the use of 2D data from cameras, radar and maps. 3D estimations leveraging both deep learning and concepts from multi-view geometry. Such accurate 3D estimations can enable human-like manoeuvres for the advanced safety of the autonomous vehicle.

• Accelerating autonomous driving solutions

Qualcomm Technologies has been at the forefront of providing connectivity solutions. Building on this solid understanding of the automotive industry, the representatives of Qualcomm are now helping drive some of the biggest innovations in AD.

Their comprehensive frameworks and tools around cross-sensory calibration, active learning, big data management, and automated verification/validation can significantly reduce the time-to-market for autonomous vehicle manufacturers.