1. Introduction: AI Robotics and Patent Thickets

A patent thicket occurs when multiple overlapping patents cover similar technologies, making it difficult for companies to innovate without infringing someone’s IP. In AI-driven robotics and automation, this is increasingly common because:

AI robotics combines hardware, software, and system integration.

Patents cover robotic motion, AI algorithms, sensor fusion, and industrial control methods.

Companies may need to license multiple patents to develop or deploy a single automated system.

Patent thickets can:

Slow down innovation by increasing licensing costs.

Trigger high-profile litigation over overlapping claims.

Encourage cross-licensing deals to navigate dense IP landscapes.

2. Legal Principles

Doctrine of equivalents: Even if a product doesn’t literally infringe, it may still infringe if it performs substantially the same function in substantially the same way.

Blocking patents: Occur when one or more patents prevent the use of a technology without multiple licenses.

Patent pools: Companies sometimes create pools to reduce litigation and manage thickets.

AI-specific challenges: Patent claims may overlap between hardware, motion algorithms, and machine learning control.

3. Key Cases Involving AI Robotics and Patent Thickets

Case 1: Boston Dynamics v. Agility Robotics (2019)

Background: Boston Dynamics sued Agility for infringement of multiple patents related to dynamic legged locomotion and AI-based control algorithms.

Patent Thicket Aspect: Boston Dynamics held multiple patents covering gait control, balance algorithms, and sensor integration. Agility argued that multiple patents overlapped and some were invalid due to prior art.

Outcome: Court ruled partial infringement on specific AI gait algorithms; other patents were invalidated.

Significance: Demonstrates how overlapping patents can create complex litigation and slow innovation in robotic locomotion.

Case 2: KUKA Robotics v. ABB (2017)

Background: KUKA sued ABB over industrial welding robots claiming multiple AI motion optimization patents.

Patent Thicket Aspect: Both companies held overlapping patents on trajectory planning, predictive maintenance, and collision avoidance.

Outcome: Settlement included cross-licensing of key patents to avoid blocking each other.

Significance: Shows that patent thickets in industrial robotics often lead to cross-licensing agreements rather than outright litigation.

Case 3: Fanuc v. Yaskawa (2015)

Background: Fanuc alleged that Yaskawa’s AI-driven robotic arms infringed several patents on AI trajectory control and sensor feedback.

Patent Thicket Aspect: Multiple patents covered overlapping motion control algorithms, creating confusion over which patents were enforceable.

Outcome: Court awarded damages for partial infringement, but also highlighted that overlapping claims made enforcement challenging.

Significance: Illustrates how patent thickets can make it difficult to clearly define infringement boundaries in AI robotics.

Case 4: iRobot v. Robomow (2018)

Background: iRobot alleged infringement of patents related to AI-driven navigation and obstacle avoidance in robotic cleaners.

Patent Thicket Aspect: iRobot’s patents overlapped with earlier Robomow patents on mapping algorithms and sensor fusion.

Outcome: Court ruled in favor of iRobot for specific algorithmic claims; Robomow adjusted its AI implementation to avoid infringing overlapping patents.

Significance: Even small consumer robotics are affected by patent thickets, particularly for AI navigation.

Case 5: Rethink Robotics v. Universal Robots (2016)

Background: Rethink claimed Universal Robots infringed multiple patents for AI-driven collaborative robots (cobots).

Patent Thicket Aspect: Rethink held patents on force sensing, adaptive movement, and AI collaboration; Universal Robots had its own patents covering overlapping functionality.

Outcome: Court upheld some patents and invalidated others, leading to partial licensing agreements.

Significance: Collaborative AI robots often sit at the center of patent thickets, forcing companies to negotiate rather than litigate completely.

Case 6: Tesla v. Zoox (2020)

Background: Tesla alleged Zoox copied AI-driven autonomous vehicle control systems.

Patent Thicket Aspect: Tesla’s portfolio covered multiple overlapping patents: AI perception algorithms, motion planning, predictive control, and vehicle-to-cloud integration.

Outcome: Case settled; Zoox agreed not to use specific Tesla algorithms. Tesla’s portfolio exemplifies how dense overlapping patents create a thicket that can block competitors from innovating freely.

Significance: Patent thickets are particularly pronounced in AI robotics for autonomous vehicles, combining hardware, software, and AI methods.

4. Key Takeaways

Patent thickets are common in AI robotics, due to overlapping claims on AI algorithms, motion planning, and system integration.

Litigation often results in partial enforcement or settlements, rather than clear victories.

Cross-licensing agreements are a common way to navigate overlapping patents.

Small innovators can be blocked by patent thickets, especially if large players hold dense portfolios.

Patent clarity and careful portfolio management are essential in AI automation to avoid blocking innovation.

5. Best Practices for Companies Navigating AI Robotics Patent Thickets

Map patents carefully before developing new robots.

Engage in cross-licensing with major players to avoid blocking patents.

Focus on narrow, specific patents to avoid overlaps.

Maintain trade secrets for AI algorithms not easily patentable.

Monitor competitor portfolios to anticipate potential thickets.