AEye, Inc. has introduced 4Sight, a new sensor family built on its (Intelligent Detection and Ranging (iDAR) platform. Countering previous assumptions that high-performing long-range 1550nm LiDAR could not achieve both solid state reliability and lower cost, 4Sight delivers performance, reliability and price.
Current LiDAR sensors rely on an array of independent sensors that collectively produce a tremendous amount of data. This requires lengthy processing time and massive computing power to collect and assemble data sets by aligning, analyzing, correcting, down sampling, and translating them into actionable information that can be used to safely guide the vehicle.
In its first instantiation, iDAR takes solid-state agile LiDAR, fuses it with a low-light HD camera, then integrates artificial intelligence to create a smart, software definable platform that enables faster, more accurate, and more reliable artificial perception.
The first 4Sight sensor to be released is the 4Sight M, available in July. The 4Sight M is designed to meet the diverse range of performance and functional requirements to power autonomous and partially automated applications.
The 4Sight family of sensors has been developed and tested over the last 18 months in conjunction with a wide range of customers and integrators in automotive, trucking, transit, construction, rail, intelligent traffic systems (ITS), aerospace and defense markets.
4Sight leverages the complete iDAR software platform, which incorporates an upgraded visualizer (which allows you to model various shot patterns) and a comprehensive SDK so it is fully extensible and customizable.
Some of the unique features of the 4Sight M are:
Software definable range optimization of up to 1,000 meters (eye- and camera-safe)
Up to 4 million points per second with horizontal and vertical resolution less than 0.1°
Instantaneous addressable resolution of 0.025°
Library of functionally-safe deterministic scan patterns that can be customized and fixed or triggered to adjust to changing environments (highway, urban, weather, etc.).
Integrated automotive camera, boresight aligned with AEye’s agile LiDAR—instantaneously generating true color point clouds. Parallel camera-only feed can provide cost-effective redundant camera sensor.
Enhanced ground plane detection to determine topology at extended ranges.
Advanced Vision Capabilities
Detection and classification of objects with advanced perception features such as intraframe radial and lateral velocity.
Detection through foliage and adverse weather conditions such as rain and fog through the use of dynamic range, full-waveform processing of multiple returns.
Detection of pedestrians at over 200 meters.
Detection of small, low reflective objects such as tire fragments, bricks or other road debris (10x50cm at 10% reflectivity) at ranges of over 120 meters.
Shock and Vibration – designed and tested for solid-state performance and reliability. 4Sight has proven in third-party testing to sustain mechanical shock of over 50G, random vibration over 12Grms (5-2000Hz), and sustained vibration of over 3G for each axis.
Automotive-qualified supply chain utilizing standard production processes and overseen by global manufacturing partners.
Designed for manufacturability using a simple solid-state architecture consisting of only 1 scanner, 1 laser, 1 receiver, and 1 SoC.
Common hardware architecture and software/data structures across all fully autonomous to partially automated applications (ADAS)—leveraging R&D and economies of scale.
4Sight can be configured for high-volume Mobility applications with SOP 2021 at an estimated 2x-5x lower price than any other high-performance LiDAR, additionally for ADAS applications with SOP 2023 4Sight is designed to be priced 1.5x-3x lower than any other long or medium range LiDAR.
4Sight series production packaging options include roof, grill, and behind the windshield, software optimized depending on the placement.
The 4Sight M is also extremely power efficient. 4Sight’s system attributes such as time-to-detect, agile scanning, shot energy optimization, power optimization of returns, and boresight camera/LiDAR data generation, greatly reduces the processing and bandwidth typically required for full perception stacks. AEye expects 4Sight sensors, when implemented in a vehicle, to be nearly power neutral in relation to a vehicle’s perception stack power budget.