According to a recent LinkedIn post from RoboSense (2498.HK), the company used its 2026 Tech Day to present a new EOCENE SPAD-SoC architecture and two flagship chipsets, Phoenix and Peacock. The post suggests these developments are intended to push LiDAR toward image-grade, high-definition 3D perception with a focus on scalability and cost structure.
Claim 55% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The EOCENE platform is described as a self-developed architecture built on a 28nm automotive-grade process that targets high sensitivity, large compute capacity and strong noise and crosstalk suppression. For investors, this indicates an effort to build a proprietary semiconductor foundation that could enhance RoboSense’s vertical integration, potentially improving margins and defensibility in automotive and robotics LiDAR.
The Phoenix chipset is portrayed as an automotive-grade monolithic SPAD-SoC offering 2,160-beam image-grade LiDAR, ultra-high resolution, 600-meter detection range and AEC-Q100 certification. The post also mentions design wins with unnamed global automakers and plans for mass production in 2026, which, if realized, could translate into future series-production revenue and strengthen RoboSense’s position in advanced driver-assistance and autonomous driving programs.
Peacock is presented as a fully solid-state SPAD-SoC targeting robotics and industrial use cases, with high-density resolution, a wide field of view and millimeter-level precision, and is planned for mass production in Q3 2026. This suggests RoboSense is seeking to diversify beyond automotive into broader automation, safety and AI-related sensing markets, potentially widening its addressable market and reducing reliance on a single end sector.
The post further previews an RGBD sensor targeted for release by the end of 2027, combining color and dense spatial data for so-called physical AI applications. Strategically, this roadmap signals RoboSense’s intention to compete not just as a LiDAR hardware supplier but as a provider of advanced perception solutions as the industry transitions to digital architectures and higher beam-count systems.
If execution and customer adoption align with the timelines mentioned, the technology shift described could support higher content per vehicle or robot and longer-term revenue visibility. However, investors may also weigh risks around commercialization, competitive responses from other LiDAR and semiconductor vendors, and the capital requirements associated with scaling new chip platforms into mass production.