Bell Textron Canada Ltd v. SZ DJI Technology Co Ltd

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Bell Textron Canada Ltd. (Canada) and Bell Textron Inc. (Delaware)
  • Defendant: SZ DJI Technology Co., Ltd (China)
  • Plaintiff’s Counsel: Baker Botts L.L.P.
• Case Identification: 4:23-cv-00722, N.D. Tex., 07/14/2023
• Venue Allegations: Venue is alleged to be proper because the Defendant is a foreign entity.
• Core Dispute: Plaintiff seeks a declaratory judgment that its products do not infringe two of the Defendant’s patents related to flight control systems for unmanned aerial vehicles.
• Technical Context: The patents relate to foundational control systems for unmanned aerial vehicles (UAVs), specifically methods for ensuring stable takeoff and for precise attitude control during flight.
• Key Procedural History: The complaint notes that the Defendant, DJI, previously filed a lawsuit in the Southern District of New York asserting infringement of the same two patents against the Plaintiff, but later voluntarily dismissed that action without prejudice. The complaint also alleges that DJI sent a letter indicating its intent to file a new infringement lawsuit, creating the "actual controversy" required for this declaratory judgment action.

Case Timeline

Date	Event
2014-03-27	U.S. Patent No. 9,126,693 Priority Date
2014-05-30	U.S. Patent No. 9,958,874 Priority Date
2015-09-08	U.S. Patent No. 9,126,693 Issued
2018-05-01	U.S. Patent No. 9,958,874 Issued
2023-03-29	Prior Lawsuit Filed by DJI in S.D.N.Y.
2023-07-13	DJI Letter Indicating Intent to Sue
2023-07-14	Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 9,126,693 - "Assisted Takeoff"

The Invention Explained

• Problem Addressed: The patent describes how traditional feedback control systems, particularly those using proportional-integral-derivative (PID) logic, can cause instability when an aerial vehicle takes off from a surface (’693 Patent, col. 1:18-28). The "memory effects of integration" combined with ground forces can lead to incorrect control inputs, increasing the risk of the vehicle tipping over or crashing ('693 Patent, col. 5:62-col. 6:5).
• The Patented Solution: The invention proposes a two-stage takeoff method. The vehicle uses a first control scheme during the initial takeoff phase, which differs from the normal flight control scheme by, for example, eliminating the integral control component ('693 Patent, col. 1:39-44). The system switches to a second, standard control scheme for normal flight only after it determines a "takeoff threshold" has been met. Critically, this determination is made using only on-board sensors that measure factors like motor output or the vehicle's own acceleration, without relying on external signals like GPS or reflected signals like those from an altimeter ('693 Patent, col. 2:34-36; col. 2:47-58).
• Technical Importance: This approach aims to provide a more reliable and smooth takeoff, particularly for novice users, by creating a self-contained system that does not depend on potentially unavailable or unreliable external signals to manage the critical transition from ground to air.

Key Claims at a Glance

• The complaint seeks a declaratory judgment of non-infringement of any of the patent's claims (Compl., Prayer for Relief ¶A). The key independent claims are Claim 1 (a method) and Claim 18 (a system).
• Independent Claim 1 recites a method comprising:
  • Increasing output to an actuator under a first control scheme.
  • Determining if a "takeoff threshold" is met based on actuator output or vehicle velocity/acceleration, where this determination is performed without using signals from an external source or signals reflected back to the vehicle.
  • Controlling the actuator using a second control scheme after the threshold is met.
• Independent Claim 18 recites a system with an actuator and a processor configured to perform the functions of the method in Claim 1.

U.S. Patent No. 9,958,874 - "Aircraft Attitude Control Methods"

The Invention Explained

• Problem Addressed: The patent identifies shortcomings in conventional cascaded PID controllers for managing a UAV's attitude (its orientation in pitch, roll, and yaw). It states these conventional methods are complex and lengthy to tune, and they are reactive, only making adjustments after a disturbance has already affected the aircraft's angular velocity (’874 Patent, col. 1:12-29).
• The Patented Solution: The invention proposes a more advanced control system that proactively incorporates the aircraft's specific physical characteristics—such as its moment of inertia, weight, and dimensions—into the control calculations ('874 Patent, col. 2:39-44). A key aspect of the solution is the introduction of an "angular acceleration loop with angular acceleration feedback" and "direct feedforward calculation," which allows the controller to anticipate and directly counteract disturbances, thereby improving response time and stability ('874 Patent, col. 7:8-13; col. 8:5-13).
• Technical Importance: By creating a control system that is customized to the specific physics of the aircraft it is controlling, the invention seeks to provide more robust and responsive attitude control than generic, reactively-tuned PID systems.

Key Claims at a Glance

• The complaint seeks a declaratory judgment of non-infringement of any of the patent's claims (Compl., Prayer for Relief ¶A). The key independent claims are Claims 1, 15, and 25 (all method claims).
• Independent Claim 1 recites a method comprising:
  • Calculating "aircraft configuration parameters" (e.g., moment of inertia) based on physical characteristics.
  • Generating a command signal based on a target attitude and these calculated parameters.
  • Using a feedback control scheme that includes an "angular acceleration loop with angular acceleration feedback."
• Independent Claim 15 is similar but focuses on assessing a "non-linear relationship between thrust of an actuator and actuator output" and using that relationship in the control scheme.
• Independent Claim 25 recites a method using a feedback control scheme that includes both an "angular acceleration loop" and a "direct feedforward calculation based on a target acceleration."

III. The Accused Instrumentality

The complaint does not identify any specific accused products, methods, or services. It refers generally to "products offered for sale and sold by Plaintiffs" (Compl. ¶¶ 12, 16). The complaint does not provide sufficient detail for analysis of the functionality or market context of any accused instrumentality.

No probative visual evidence provided in complaint.

IV. Analysis of Infringement Allegations

The complaint is for a declaratory judgment of non-infringement and therefore does not contain affirmative infringement allegations or a claim chart. Instead, it makes the conclusory statements that "No claim of the ’693 Patent is infringed by Plaintiffs" (Compl. ¶14) and "No claim of the ’874 Patent is infringed by Plaintiffs" (Compl. ¶18). In the absence of specific infringement theories from the patentee (DJI), the following points of contention are identified based on the patent claims and the nature of the technology.

• Identified Points of Contention:
  • ’693 Patent - Technical Questions: A central question will be how the Plaintiff's (Bell's) flight controllers manage the takeoff sequence. Do they use a single control scheme for both takeoff and flight, or a multi-stage scheme? If multi-stage, what data is used to trigger the transition? The analysis will focus on whether the transition trigger relies exclusively on on-board, non-reflected data (e.g., internal inertial measurement unit acceleration, motor command values) as the claims require, or if it uses data from external sources (e.g., GPS-derived altitude) or reflected signals (e.g., a radar or ultrasonic altimeter), which would place it outside the claim scope.
  • ’874 Patent - Scope & Technical Questions: The dispute will likely focus on the architecture of the Plaintiff's attitude control software. Does the software explicitly calculate and use "aircraft configuration parameters" like moment of inertia as a direct input to its control algorithms? Does it implement a distinct "angular acceleration loop" and "direct feedforward calculation" as described in the patent? Or, does it use a more conventional cascaded PID controller that may achieve a similar functional result without meeting the specific structural and method limitations of the claims?

V. Key Claim Terms for Construction

’693 Patent

• The Term: "without (1) using signals from a source external to the aerial vehicle, or (2) using signals reflected to the aerial vehicle" (Claim 1)
• Context and Importance: This negative limitation is central to distinguishing the claimed invention from systems that might use GPS, barometric altimeters, or radar/ultrasonic altimeters to determine when takeoff is complete. The entire non-infringement case for this patent could hinge on whether the Plaintiff's system uses any such signals for this purpose.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation (favors non-infringement): The specification consistently emphasizes the self-contained nature of the determination. It states the system "need not receive any signals from external devices, such as global positioning system (GPS) satellites" ('693 Patent, col. 11:50-55) and that the determination is made "without requiring any signals from outside the aerial vehicle" ('693 Patent, col. 6:32-34). This supports an interpretation that the use of any such external or reflected signal in the takeoff threshold logic would avoid infringement.
  • Evidence for a Narrower Interpretation (favors infringement): A patentee might argue that this limitation only applies to the primary or determinative signal, and that ancillary use of external data for other purposes, or as a secondary check, does not vitiate the element. However, the claim language "without... using" is absolute and provides little textual support for such a narrow reading.

’874 Patent

• The Term: "angular acceleration loop with angular acceleration feedback" (Claims 1, 15, 25)
• Context and Importance: Practitioners may focus on this term because it appears to be the core technical innovation intended to separate the invention from prior art cascaded PID controllers. The outcome of the case will likely depend on whether the Plaintiff's controller has a software architecture that meets this definition.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation (favors non-infringement): A party accused of infringement would argue this requires a specific, explicitly implemented control loop where angular acceleration is the primary controlled variable being fed back, as distinct from the derivative term of an angular velocity PID controller. The specification describes this loop as being added to a PID scheme and acting as "direct control" to "directly suppress disturbances" ('874 Patent, col. 7:8-13; col. 20:57-61). This suggests a specific structure, not just a functional result.
  • Evidence for a Narrower Interpretation (favors infringement): The patentee would likely argue for a functional definition, where any control system that calculates and uses angular acceleration to adjust its output falls within the claim's scope, regardless of the specific software implementation. They may point to language stating the loop "can enhance dynamic tracking performance and disturbance resistance" ('874 Patent, col. 28:35-36) as evidence that the function, not the form, is what is patented.

VI. Other Allegations

• Indirect Infringement: The complaint does not contain allegations related to indirect infringement.
• Willful Infringement: The complaint does not contain allegations related to willful infringement.

VII. Analyst’s Conclusion: Key Questions for the Case

• A primary issue will be one of technical architecture: The case for both patents will depend heavily on discovery into the source code and design of the Plaintiff's flight control systems. For the ’693 patent, what are the precise data inputs for its takeoff-to-flight mode transition logic? For the ’874 patent, does its attitude controller explicitly calculate physical parameters and implement a distinct "angular acceleration loop," or does it utilize a conventional control design?
• A second core issue will be one of claim construction: The dispute over the ’693 patent will turn on the breadth of the negative limitation "without... signals from a source external... or... reflected." The viability of the ’874 patent allegations will depend on whether "angular acceleration loop" is construed to require a specific software structure or whether it can be met by any system that functionally accounts for angular acceleration.