Shenzhen-based RoboSense has recently announced the start of an automotive-grade solid-state LiDAR production line in China. The advanced LiDAR solution manufacturer has also joined hands with Inceptio Technology and will be powering its self driving system.
Season Wang, Vice President, Brand Ecology Strategy and Co-partner at RoboSense, spoke to Auto Futures about the company’s newest developments.
“The biggest challenge for autonomous driving is safety. RoboSense’s mission is to possess outstanding hardware and artificial intelligence capabilities to provide smart solutions that enable robots (including vehicles) to have perception capability more superior to humans. Moreover, we saw the potential growth in autonomous driving especially in passenger cars, for which, LiDAR is the important and indispensable sensor for ADAS and AD system. Therefore, we decided to develop a LiDAR product that meets all the industry requirements including automotive-grade quality and mass-production capability,” says Wang.
To date, RoboSense’s LiDAR systems have been applied to future mobility, including autonomous driving passenger cars, such as RoboTaxis, RoboTrucks and RoboBuses. Its automated logistics vehicles have been used by domestic and international autonomous driving technology companies, OEMs, and Tier1 suppliers.
“The RS-LiDAR-M1 adopts MEMS scanning technology reduces the number of parts of LiDAR (from hundreds to dozens), greatly lowering the cost, shortening the production time, and achieving breakthroughs in manufacturability and reliability of LiDAR products. In December last year, the RS-LiDAR-M1 samples were shipped to an OEM customer in batches and became the world’s first mass-production-ready automotive-grade MEMS solid-state LiDAR delivered,” he explains.
“MEMS Smart LiDAR Sensor (RS-LiDAR-M1), which is the world’s first and smallest MEMS Smart LiDAR sensor incorporating LiDAR hardware, built-in dedicated SoC and AI sensing algorithm can transform conventional 3D LiDAR sensors into full data analysis and comprehension systems. It speeds up the environmental information processing for self-driving car’s decision making, and not only helps OEMs to release the R&D pressure and makes it safer for humans, but also reduces the requirements to the vehicle’s ECU and the cost of the system, saving the cost needed by OEMs to develop their own AI sensing algorithm.”
“RoboSense has been developing the RS-LiDAR- Algorithm for more than 10 years, which has been verified by over 100 global partners in various types of autopilot scenarios. After extensive data training and model optimization, both algorithm performance and stability have been proven far superior,” adds Wang.
Scanning gets Revolutionary
I was keen to know more about RoboSense’s automotive-grade LiDAR, what made it special and what more benefits does it possess for the autonomous mobility space.
To help me understand, Wang says: “The mechanical LiDAR, whose performance improvement relies heavily on stacking components, was inevitably disadvantaged by its large size, poor manufacturability, poor reliability, and high cost. To achieve large-scale commercialization of autonomous vehicles, a new generation of automotive-grade high-precision 3D environment sensing LiDAR is required to fulfil the industry’s strict requirements. This includes, but is not limited to, high performance, high reliability, small size and easy scalability, low cost, and high manufacturability.
“Our LiDAR adopts MEMS scanning technology, which is the first solution that has successfully gotten rid of bulky motors. Its core component, the MEMS micro-galvanometer, uses mature semi-conductor technology with a high level of integration. It can achieve the equivalent scanning effect of a 100-beam mechanical LiDAR, with just one pair of transceiver modules. With this revolutionary scanning method, the MEMS LiDAR transceiver components that are required are fewer from hundreds to dozens in comparison to traditional mechanical LiDARs, greatly reducing the cost and shortening production time – achieving a breakthrough in manufacturability.
“For automotive LiDAR, it usually takes several years for any new platform to move from a concept to a truly stable mass-produced product. Over five years of R&D investment, the automotive-grade solid-state MEMS LiDAR RS-LiDAR-M1 has steadily moved from concept to its final mass production readiness stage. in January 2021, RoboSense has officially unveiled the SOP version of the automotive-grade LiDAR RS-LiDAR-M1, which features a slim design, excellent performance, and reliable point cloud quality,” he explains.
The automotive-grade solid-state LiDAR production line is designed for the mass production of LiDAR.
Since July 2020, the RS-LiDAR-M1 has received many orders for mass-production vehicle models around the world. The first customer was an OEM from North America.
In December 2020, the RS-LiDAR-M1 samples were shipped to this OEM customer in batches, which made RoboSense the first company to deliver the automotive-grade solid-state LiDAR in batches to OEM customers and proved M1 as the first mass-production-ready automotive solid-state LiDAR.
That’s not all, RoboSense recently announced the release of a new version of its Sensor Evaluation System for Autonomous Driving with added features.
When asked what this means for customers, Wang replies: “The RS-Reference 2.1 can not only evaluate the fusion results of intelligent driving perception fusion system, but also provide dedicated evaluation tool modules based on different types of sensors such as LiDAR, millimeter-wave radar, and camera. Moreover, customized tool modules can be developed according to customers’ requirements to carry out in-depth analysis on the performance of the perception system.
“The RS-Reference system includes the RoboSense 128-beam LiDAR RS-Ruby, Leopard camera, Continental 408 millimeter-wave radar, GI-6695 RTK, and has added two RoboSense RS-Bpearl LiDAR for near-field blind spots detection in the 2.1 version. The whole system can achieve full coverage and perception redundancy, reliability, and accuracy of original environment data.”
“The massive ground truth data of real scenes generated by the RS-Reference can be used to construct simulation scenes. In this way, not only the simulation scenes can get a reliable source of data, but also the authenticity of the data is greatly improved. Since its launch, the RS-Reference has been used by many top OEMs and Tier1s globally to solve sensing problems of many smart driving projects, and has won high recognition and a large number of orders from customers,” he explains.
One recent development for the company was the cooperation with Inceptio in 2019 to provide high-performance automotive-grade MEMS solid-state LiDAR RS-LiDAR-M1 to assist with the ‘Xuanyuan’ system, the first full-stack self-developed autonomous trucking system for mass production.
The two companies announced the partnership on the mass-produced L3 heavy-duty freight truck equipped with ‘Xuanyuan’ system, to be launched on the market by the end of 2021.
Further news was about China’s first ever automotive-grade solid-state LiDAR production line.
I asked Wang to tell me more about the facility and he told me: “The automotive-grade solid-state LiDAR production line is designed for the mass production of LiDAR, to help autonomous vehicles achieve mass production. Thanks to the integrated and compact design of the M1, the production process is greatly simplified, and the automotive-grade production line can be quickly expanded to catch up with the increased order volumes. RoboSense’s automotive-grade production line covers 4 major parts of the solid-state LiDAR M1 assembly, including the receipt and management of components and materials, SMT and bonding of PCBA boards, overall assembly and light adjustment, calibration and factory testing.
“The automotive-grade production line realizes full-process automation, which greatly improves production efficiency, and ensures product performance consistency and information traceability.”
For the future, RoboSense will focus on accelerating automotive-grade mass-production of the solid-state M1 product as the priority, to help more OEM to achieve the goal of mass production of self-driving vehicle models, and push for on-vehicle tests for technology optimisation.
“Focusing on all autonomous driving applications and intelligent roads, we will continue to improve the performance and cost of the mechanical LiDAR product line-up. At the same time, we are also researching the future LiDAR technology such as Flash and OPA,” concludes Wang.