Mapping, Detection and Localisation System

Technology Overview

The perception technology provides environment perception technology to ensure the ego vehicle can detect, comprehend, and make real-time decisions to navigate autonomously in the real-world environment. This technology utilizes various 3D visual technologies to create a real-time understanding of the environment, assisting in decision support for the vehicle’s next course of action.

The 3D technologies include a stereoscopic view for depth estimation and 3D LIDAR point cloud information on top of other sensors such as IMU, GPS, Radar and ultrasonic to enhance environment perception and decision support.

The perception technology is designed for advanced environmental understanding, emphasising localisation, mapping, and cognitive functions such as object detection and recognition.

Key Features

The perception technology provides the following features:

• Sensor Fusion for Multi-Sensory Inputs

  A comprehensive sensor fusion platform that seamlessly combines inputs from multiple sensors, including LIDAR, Stereo camera, GNSS, IMU, Radar, and ultrasonics.

• Data Fusion

  A data fusion capability for environment mapping and localization, utilizing inputs from both stereoscopic camera views and 3D LIDAR data.

• Obstacles Detection

  Enhanced obstacle recognition and comprehensive
  environment understanding, serving both navigation and mission-specific objectives.

Technology Benefits

• Accurate Positioning Indoors and Outdoors

  The perception technology provides accurate positioning and localization in both indoor and outdoor environment.

• Flexible Support of Sensors

  The sensor fusion platform allows for flexible sensor configurations, accommodating various types of sensors to facilitate customization for diverse purposes.

The perception technology is centered around environment understanding, particularly in localization, mapping, and cognitive aspects like object detection and recognition. It processes multiple sensory inputs, such as camera views, LIDAR, and positioning sensors, on edge devices to determine the relative location of objects within the digital mapped and actual environment.