Hesai says its 6D full-color lidar merges color and depth on a single chip for autonomous driving
Hesai Technology is pitching its 6D full-color lidar, built on the Picasso ASIC platform unveiled in April, as one sensor that delivers chip-aligned color and depth for autonomous driving and Physical AI.
- 6D — RGB color plus 3D position and distance
- April 2026
What Happened
Hesai Technology argues that as autonomous driving and embodied intelligence expand, demand is growing for high-quality, high-confidence ground-truth data, and that 6D full-color lidar is emerging as the next-generation sensor to supply it. In April 2026 the company unveiled Picasso, which it calls the world's first 6D full-color ultra-sensitive lidar ASIC platform, achieving native pixel-level alignment of high-precision 3D spatial information and high-quality RGB color data at the chip level. Hesai says this produces inherently consistent, estimation-free ground-truth data, eliminating the decision-making hesitation caused by sensor conflicts in autonomous driving systems and positioning native RGB-D data as critical infrastructure for perception in the Physical AI era.
- Separate camera and lidar streams that must be synchronized, calibrated and fused versus one sensor carrying color, position and distance per pixel
- System level or optical level versus chip level, natively inside the Picasso SPAD-SoC
- Images and point clouds requiring registration and 3D annotation versus native RGB-D data with no stitching or reverse inference
- Traffic lights: color and spatial information are inherently aligned, so the system can identify signal states without aligning and fusing separate camera and lidar outputs.
- Rain and fog: color information can help filter noise more efficiently.
- Ambiguous objects such as a white plastic bag blown across the road: visual and spatial information no longer needs repeated cross-checking between sensors.
- Distant objects occupying only a handful of pixels, from traffic cones to children's bicycles and pets: chip-level alignment keeps color and distance matched, reducing localization errors from image-to-point-cloud misalignment.
Hesai links the sensor to the data bottleneck in AI model training, noting that NVIDIA has highlighted the limited availability, high collection costs and difficulty of scaling real-world robotics data. Traditional RGB-D data requires separate collection of images and point clouds followed by time synchronization, extrinsic calibration, data registration and 3D annotation, a pipeline Hesai describes as long, costly and prone to errors during conversion and registration. Picasso SPAD-SoC integrates RGB sensing and time-of-flight ranging at the chip level, so every point the lidar outputs natively carries both RGB and XYZ information with chip-level spatiotemporal alignment. Hesai says this gives Physical AI direct access to native RGB-D data, enabling automated annotation, shortening the path from collection to model training, improving data-production and training efficiency and reducing the cost of training at scale.
Hesai also frames 6D lidar as an answer to the industry's habit of stacking more sensors to improve perception, which brings extra hardware, wiring harnesses and interfaces plus added complexity in time synchronization, calibration, data registration and fusion computation. Because the 6D sensor captures distance, position, color and other multi-dimensional information in one pass, Hesai says it can take on some of the visual information previously provided by standalone cameras. That would reduce the number of sensors and shorten the cross-sensor fusion pipeline, simplifying system architecture and lowering hardware, compute and development costs, a shift Hesai describes as moving from multi-sensor fusion toward a more integrated perception architecture.
Previously from Hesai Technology
Earlier in October 2026, Hesai launched Kosmo, a 3D world data engine that pairs an 8K AI spatial camera with a cloud platform to turn real places into 3D training data for robots and world models. In September 2026 the company published a plain-language guide to lidar explaining how laser pulses become 3D point clouds, why lidar still works in darkness, and where the technology goes next, including full-color sensing.
- Hesai launches Kosmo 3D world data engine with AI spatial camera for robot training
- Hesai explains how lidar 3D perception works, from point clouds to full-color sensing
Background drawn from MotorClaw's earlier coverage of Hesai Technology's official releases.
Why this matters
If color and distance are truly aligned inside the chip, autonomous driving systems would no longer have to reconcile separate camera and lidar streams, which Hesai says causes conflicts, mapping errors and hesitation in decision-making. Hesai also argues a single 6D sensor could absorb some of the visual work of standalone cameras, cutting sensor count and lowering hardware, compute and development costs for carmakers, robotics developers and AI model trainers.
Terms in This Story
- lidar
- A sensor that measures distance by firing laser pulses and timing their return, building a 3D map of the surroundings.
- time-of-flight (ToF)
- A ranging method that calculates distance from how long a light signal takes to travel to an object and back.
- RGB-D
- Data that pairs color information (red, green, blue) with depth, so each pixel carries both appearance and distance.
- ASIC
- An application-specific integrated circuit, a chip designed for one dedicated task rather than general-purpose computing.
Summarised from the linked release; details can be imperfect — always verify against the original source.