Autonomous Driving Patent War Warms Up with LiDAR Patent Surviving the PTAB

June 30, 2019

A “Laser Imaging Detection and Ranging” (LiDAR) device is a critical sensor for many autonomous driving solutions.   In Quanergy v. Velodyne, IPR2018-00255 (PTAB, May 23, 2019), one of Velodyne’s LiDAR patents survived the invalidity challenge mounted by its rival Quanergy.  The IPR petition in this matter was filed prior to the USPTO’s recent implementation of the Phillips claim interpretation standard for IPRs.  Nevertheless, even under the “broadest reasonable interpretation” standard, the PTAB found that the intrinsic evidence sufficiently limited the scope of the term “LiDAR” to allow the patentee to effectively distinguish the prior art.

The challenged patent, U.S. Pat. No. 7,969,558, relates to a LiDAR based 3-D point cloud measuring system.  The system has two assemblies of 32 pairs of pulsed laser emitters and photodiode detectors.  The petitioner challenged the claims of the ’558 patent as obvious under 35 U.S.C. § 103 over Mizuno (Japanese Unexamined Patent Appln. Pub. No. H3-6407, pub. Jan. 11, 1991), and a combination of Mizuno and Kilpela (“Precise pulsed time-of-flight laser range finder for industrial distance measurements,” REVIEW OF SCIENTIFIC INSTRUMENTS, Vol. 72, No. 4, pp. 2197–2202 (Apr. 2001).  

The patent owner argued that “LiDAR” should be interpreted to mean “only pulsed ToF [time-of-flight] lidar” that “excludes triangulation and other techniques not analogous to those used in radar.”  The petitioner disagreed, arguing that the term “LiDAR” should be broadly interpreted as including triangulation and other techniques as well, relying mainly on extrinsic evidence.  The extrinsic evidence included the petitioner’s expert opinions along with several papers cited in support of those opinions. However, many of the papers were published years after the relevant time frame of the invention. And the only paper contemporaneous to the invention drew a clear distinction between “triangulation-based” sensors and “time-of-flight” (TOF) sensors.  After considering the intrinsic evidence, the PTAB sided with the patent owner.  The PTAB found that the specification focuses exclusively on the pulsed ToF LiDAR, and no other types.  This interpretation was critical to the ultimate finding of non-obviousness.

The PTAB next discussed Mizuno, finding that it does not teach or suggest a pulsed ToF LiDAR system and therefore does not teach LiDAR within the meaning of the disputed claims.  In fact, even the petitioner’s expert admitted that Mizuno’s system is a “very simplistic system,” and that it was not based on ToF.  Attempting to fill in the gap, the petitioner added another reference, Berkovic (Garry Berkovic, Optical Methods for Distance and Displacement Measurements, ADVANCES IN OPTICS AND PHOTONICS (2012)), arguing that Berkovic shows “it was well known to use pulsed ToF techniques to measure objects at close distances, such as less than 1 m.”  The PTAB disagreed:  Berkovic teaches that the ToF technology works best for measuring objects at distances greater than 50 meters, but calls into question the accuracy of pulsed ToF LiDAR for short-range measurement.  Thus, it would not have been obvious to use pulsed ToF LiDAR in a short-range measuring device like Mizuno.  Further, the petitioner’s reliance on Kilpela to fill in the gap of Mizuno failed as well, because Kilpela also discloses an increasing rate of error and inaccuracy using the pulsed ToF device at short-range distances.

The PTAB further supported its finding of non-obviousness based on “secondary indicia of non-obviousness,” specifically, the patent owner’s evidence of long-felt need, industry praise, and commercial success.  For example, five out of six top finishing vehicles in a 2007 DARPA challenge were equipped with Velodyne’s patented LiDAR sensors.  Forbes labeled Velodyne as the top supplier of advanced automotive LiDAR.  And Velodyne generated hundreds of millions of dollars in revenue from 2007-2018 from sales of the patented sensors.

The PTAB held that the petitioner did not prove by preponderance of evidence that the challenged claims of the ’558 patent were unpatentable.  This case demonstrates the continued importance of the intrinsic evidence, such as the specification, for claim construction.  Even under a broadest reasonable interpretation standard, the interpretation must be reasonable in light of the specification.  A patent drafter should describe the invention’s technology clearly and thoughtfully in the patent application to support favorable interpretations in later disputes.

Author: Ray Huang