Flexible deep brain neural interfaces, as an important research direction in the field of neural engineering, have broad application prospects in areas such as neural signal detection, treatment of neurological diseases, and intelligent control systems. However, chronic inflammatory responses caused by long-term implantation and the resulting electrode failure seriously hinder the clinical development of this technology. This review systematically explores the long-term stability issues of flexible deep brain neural interfaces, with a focus on analyzing the synergistic optimization of electrode geometric morphology and implantation strategies in regulating inflammatory responses. Additionally, this paper delves into innovative strategies, such as passive enhancement of biocompatibility through electrode surface functionalization and active inhibition of inflammation through drug-controlled release systems, offering new technical paths to extend electrode lifespan. By integrating and reviewing existing innovative methods for deep brain flexible electrodes, this study provides an important theoretical foundation and technical guidance for the development of high-stability neural interface devices.

Les architectures de véhicules à gestion logicielle (SDV) devraient dominer les ventes de véhicules neufs ; les projections indiquent que d’ici 2035, plus de 75 % des voitures neuves seront construites selon les principes SDV. Actuellement, tous les grands constructeurs automobiles mondiaux investissent massivement dans les programmes SDV. À mesure que la valeur des véhicules repose de plus en plus sur les logiciels, les semi-conducteurs (silicium) deviennent également essentiels, entraînant une transformation profonde de la puissance de calcul, de la gestion de l’énergie et de la quantité totale de semi-conducteurs par véhicule, qui pourrait potentiellement doubler à terme. Parmi les principaux facteurs de cette croissance figurent la centralisation de l’architecture électronique/électrique (E/E) – qui augmente structurellement la quantité de silicium et de logiciels par véhicule – et l’adoption des systèmes avancés d’aide à la conduite (ADAS) de niveau 2+/3, qui génère une demande exponentielle de plateformes informatiques évolutives, performantes et certifiées conformes aux normes de sécurité.

The automotive industry is going through one of its biggest transformations in history. For over a century, cars were defined by engines, horsepower, and mechanical innovation. Today, that definition is changing rapidly. The future of mobility is no longer driven by hardware alone it is powered by code. This shift has given rise to a new concept known as Software-Defined Vehicles. Software-Defined Vehicles are not just cars with software features; they are vehicles where software controls most of the functions, from performance and safety to entertainment and user experience. In simple terms, the car is becoming a digital platform on wheels.

Is China really “there” already, producing cars with centralized computers and a hypervisor? Coming out of CES 2025 earlier this year, we covered a Snapdragon innovation Qualcomm was pitching to automakers, capable of running an entire car on a single chip. Now we just returned from China where we drove a similar system by Chinese EV maker Nio—which designed, developed, and manufactured its own chip in-house, along with the software that runs on it.

China's automotive industry has undergone a fundamental transformation over the past two decades, shifting from a manufacturing-focused economy to a global innovation laboratory. The pivotal change occurred when automotive competition shifted from internal-combustion engines toward batteries, power electronics, software, connectivity, and artificial intelligence—domains where Chinese companies faced fewer legacy constraints. This enabled rapid market adaptation and accelerated development cycles that traditional five-year automotive timelines cannot match. China has developed a highly integrated innovation ecosystem combining the world's largest automotive market with world-class capabilities in batteries, electronics, semiconductors, and digital services. Leading Chinese automakers exemplify this transformation: BYD leverages deep vertical integration in batteries and power electronics, Geely combines Chinese innovation speed with global assets, and Chery has developed strong international distribution capabilities. The competitive intensity within China's market forces continuous innovation, as standing still for even a year risks obsolescence. Notably, foreign automakers increasingly enter China not merely to sell vehicles but to access cutting-edge technology, suppliers, and development capabilities—reversing the traditional knowledge flow established decades ago.

UN Regulation R156 establishes a framework for managing software updates in type-approved vehicles by requiring manufacturers to identify software dependencies through RxSWIN identifiers, with one identifier per applicable regulation such as R79 for steering and R13-H for braking. When software updates modify approved components, the RxSWIN must be updated to reflect changes and determine whether type approval remains valid or requires extension. A critical compliance gap emerges at the intersection of R155 cybersecurity requirements and R156 update management: cybersecurity threat and risk assessments (TARA) are validated for specific software states, but updates transition vehicles to new states. If the TARA analysis isn't revisited for modified components, the approval becomes dependent on outdated cybersecurity arguments. Organizationally, this fragmentation occurs because separate teams manage cybersecurity management systems, update pipelines, and RxSWIN records with no designated owner. To address this, manufacturers should maintain a single comprehensive record per release documenting RxSWIN identifiers before and after updates, affected components, relevant TARA rows, and dated decisions confirming cybersecurity case validity or re-testing completion.

This comprehensive resource provides an interactive comparison of leading automotive operating systems and vehicle software platforms as of Q3 2026. The guide covers major platforms including BlackBerry QNX, Google Android Automotive OS, Toyota Arene OS, BMW Operating System X, Tesla Vehicle Software, NIO SkyOS, XPeng Tianji AIOS, and numerous others. The comparison encompasses critical technical specifications including system scope across body control, chassis, drive, automated driving, and infotainment functions; supported processor architectures such as ARM, x86, and RISC-V; and advanced features like hypervisor support and functional safety certifications up to ASIL D level. A notable addition is the dedicated AI capabilities section, distinguishing between supported models and truly integrated on-device implementations, including LLM, VLM, and VLA support with edge-cloud hybrid inference options. The resource details middleware protocols, connectivity standards, and OTA update capabilities. Sourced from official vendor documentation and industry reports, this platform serves automotive engineers, developers, and researchers evaluating embedded systems for vehicles and autonomous applications, enabling customizable analysis focused on safety certifications, real-time performance, ADAS integration, and AI model capabilities.

Autonomous freight technology is transitioning from vehicle demonstrations toward operational deployment, with the critical next phase focusing on building the technical, regulatory, and operational infrastructure to enable driverless operation. Recent announcements at IAA Transportation in Hanover exemplify this shift: Lidl and Einride launched a Level 4 cab-less battery-electric freight service on German public roads, while MAN shared insights from its Atlas-L4 programme and Chinese specialist Neolix announced European expansion plans. MAN's framework describes autonomous transport systems across three integrated layers: the physical vehicle, decision-making software, and a control tower managing broader operations. Modern trucks increasingly feature advanced driver assistance systems providing lateral and longitudinal control, while MAN's Atlas-L4 programme has developed redundant steering, braking, and electrical architectures enabling safe Level 4 operation following system failures. The significance extends beyond technological achievement to fundamental supply chain restructuring: removing the driver eliminates the organizing principle that has traditionally structured road freight economics and workflows. This transformation suggests the industry is entering genuine commercial deployment rather than experimental phases, requiring substantial infrastructure development alongside vehicle technology maturation.

Automotive manufacturers are increasingly abandoning phone mirroring applications like Apple CarPlay and Android Auto in favor of native in-car infotainment systems powered by Google's Android ecosystem. Google supports this shift through two distinct platforms: Android Automotive Operating System (AAOS) and Google Built-in, both independent operating systems that function without requiring a connected smartphone. General Motors exemplifies this transition, having phased out Android Auto since 2023 in favor of Google Built-in solutions. While manufacturers highlight enhanced seamless integration between driver and vehicle as a benefit, some consumers express skepticism, suspecting these built-in systems prioritize revenue generation through subscriptions or data monetization over improved user experience. Android Automotive distinguishes itself as a comprehensive, open-source software solution enabling manufacturers to customize their entire in-vehicle ecosystem—controlling radio, climate, cameras, mirrors, lighting, and instrument clusters. This flexibility allows automakers to develop brand-specific systems tailored to their customers' needs. Both platforms operate independently from smartphones, though connecting to the same Google account enhances compatibility. For consumers, understanding these technical differences proves important, as each platform significantly impacts how infotainment systems appear and function within vehicles.

Automakers are generating record profits through software monetization strategies that increasingly bypass traditional dealerships. General Motors exemplifies this trend, reporting record quarterly revenue despite declining vehicle sales, with connected platforms like OnStar and driver-assistance features such as Super Cruise generating over $800 million in software revenue within three months, positioning the company toward $7.5 billion in deferred subscription revenue annually. These recurring software services, bundled as standard features on high-margin vehicle trims, generate approximately 70% gross margins that flow directly to manufacturer servers without dealer participation. This structural shift fundamentally realigns lifetime customer value economics, as dealerships shoulder inventory costs, overhead, and physical preparation while manufacturers capture lucrative digital touchpoints and recurring subscription fees. The implications are significant for automotive retailers, who risk becoming secondary service providers relegated to warranty work if they lack proprietary digital engagement tools. Manufacturers employ machine learning and automated workflows to optimize administrative efficiency and maintain direct customer communication channels, widening the competitive distance between initial showroom sales and subsequent service visits. Rather than resisting manufacturer software platforms, dealerships must prioritize maintaining brand authority and customer relevance during the critical three-to-five-year intervals between vehicle purchases, as OEM applications are engineered primarily to monetize factory subscriptions rather than drive local service revenue or facilitate trade-in opportunities.

The automotive industry is transitioning toward a subscription-based model where vehicle features can be activated, upgraded, or purchased through ongoing payments, similar to digital streaming services. According to Statista's Automotive Trends 2026 report, the global market for in-vehicle digital solutions is projected to grow from $8.6 billion in 2025 to $13.9 billion by 2030, representing approximately 10% annual revenue growth based on Oliver Wyman analysis. Features including navigation, entertainment, driver assistance, telematics, and vehicle management could become part of this emerging revenue stream. The model leverages software installed during manufacturing, allowing customers to add or modify features post-purchase while manufacturers maintain commercial relationships throughout vehicle lifespans. This fundamental shift from one-time vehicle sales to recurring digital payments represents a significant business transformation for automakers, enabling dashboards and screens to function as revenue-generating platforms long after dealership purchase, creating new opportunities beyond traditional maintenance and spare parts sales.

UN Regulation No. 155, established by the UNECE World Forum for Harmonization of Vehicle Regulations, mandates cybersecurity as a prerequisite for vehicle type approval. The regulation, which entered force in January 2021 under the UNECE 1958 Agreement, applies to contracting parties including EU member states, the UK, Japan, and South Korea, though the USA, Canada, and China maintain independent approaches. The regulation imposes two primary requirements: manufacturers must obtain a Certificate of Compliance for their Cybersecurity Management System (CSMS) demonstrating organizational processes for lifecycle cybersecurity management, and each vehicle type requires approval confirming risk identification, mitigation implementation, testing, and field detection protocols. In the EU, UN R155 became mandatory through Regulation (EU) 2019/2144 for new vehicle types from July 2022 and all new vehicles from July 2024. The regulation's Annex 5 provides a comprehensive checklist covering identified threats, vulnerabilities, and mitigations both within vehicles and external systems. For original equipment manufacturers, cybersecurity documentation now constitutes part of homologation files subject to technical service review, while suppliers must provide evidence including threat and risk analyses aligned with regulatory expectations.

Toyota Motor Corporation has announced a strategic shift to transition all 50 of its vehicle models to Software-Defined Vehicles (SDVs) by approximately 2030, mirroring the smartphone model where vehicle functions receive continuous software updates post-purchase. Tesla pioneered this approach in 2012 and now operates a subscription-based autonomous driving model at $99 per month, establishing an advanced business paradigm. According to Mobility Global's SDV Readiness report from August 2026, Tesla and Chinese manufacturers occupy the highest Level 4-5 tier, while Japanese competitors including Toyota remain at Level 1, with European players like Volvo and BMW advancing progressively. The disparity reflects fundamental manufacturing philosophies: Japanese automakers traditionally relied on 'suriawase' development with meticulous part-by-part refinement, whereas SDVs demand vertical integration where software and major components are controlled in-house—a strength demonstrated by Tesla and BYD. The transition imposes substantial costs and time investments; Volkswagen exemplifies this shift by partnering with American and Chinese firms rather than developing automotive software independently. Toyota's conversion, given its 10 million annual vehicle sales, will significantly impact the entire supply chain. Traditional vehicles utilized approximately 100 distributed ECUs, while SDVs consolidate to fewer high-performance central computers. BYD has surged to third place globally in ECU sales at $14.9 billion, surpassing Continental, with emerging Chinese players like Desay SV gaining prominence. The absence of Japanese companies in top ECU rankings highlights technological concerns, prompting Honda and Nissan to pursue joint development partnerships.

As automakers accelerate software-defined vehicle (SDV) deployment, cybersecurity vulnerabilities pose escalating risks. SDVs' extensive connectivity—enabling route planning, over-the-air updates, and V2X communication—creates multiple attack vectors across their 100+ million lines of code and 100 electronic control units. A recent demonstration exposed critical weaknesses when a cybersecurity researcher gained unauthorized control of a BYD Shark 6's headlights, wipers, door locks, infotainment, and microphone without passwords, potentially capturing sensitive driver information. Cyberattacks cost the automotive industry $22.5 billion in 2024, representing a 22-fold increase from 2022, encompassing ransomware, data breaches, and system downtime. Beyond financial losses, security breaches erode consumer trust and brand reputation. To address these threats, the industry has implemented ISO/SAE 21434, an international engineering standard for OEMs and suppliers, though compliance remains optional. Binding regulations now mandate protection: UNECE R155 and R156 for EU, UK, Japan, and South Korea, while China enforces GB 44495, establishing mandatory cybersecurity requirements for vehicle manufacturers in these regions.

Toyota, Woven by Toyota, Denso, Renesas, Qualcomm, and NVIDIA have jointly unveiled an ambitious blueprint to transform approximately 50 Toyota vehicle models into software-defined vehicles (SDVs) by 2030, reaching roughly ten million units annually. This represents a fundamental restructuring of automotive design, development, verification, and post-launch updates rather than incremental technological advancement. The cornerstone is Arene, a proprietary software development platform developed by Woven by Toyota, which enables continuous safety improvements, AI-assisted driving, and evolving in-vehicle digital experiences throughout a vehicle's lifecycle. The initiative integrates multiple partners' technologies: NVIDIA's DRIVE AGX Orin and DriveOS for autonomous driving, Qualcomm's Snapdragon Cockpit for in-vehicle systems, and Renesas chipsets for vehicle control. The implementation spans a four-layer architecture encompassing software tools, cloud MLOps infrastructure, and distributed vehicle computing resources. Key technical challenges include harmonizing Arene with NVIDIA's autonomous systems, managing mixed-criticality environments, implementing over-the-air updates across diverse hardware configurations, and advancing from Level 2 to Level 4 autonomous capabilities. This transformation reflects Toyota's strategic bet on vertical integration of software architecture while leveraging global supplier ecosystems.

LG Vehicle Solution has published a white paper examining the practical challenges original equipment manufacturers (OEMs) face when transitioning to software-defined vehicles (SDVs), including escalating software complexity, elevated total cost of ownership, divergent segment-specific requirements, and stringent regulatory compliance demands. The paper proposes that flexible, scalable electrical and electronic (E/E) architectures can effectively address these interconnected challenges. LG Vehicle Solution's proposed solution leverages four core capabilities: hardware virtualization, modular high-performance computing, software orchestration via their proprietary PICCOLO platform, and open ecosystem collaboration through S-CORE and SOAFEE frameworks. The analysis compares distributed, domain-oriented, and zonal E/E architectures, demonstrating how standardized software foundations and consolidated computing resources can reduce duplicated development work across vehicle programs. By consolidating computing resources and improving software portability, OEMs can accelerate over-the-air update readiness while managing costs. This integrated approach enables manufacturers to simultaneously coordinate infotainment, connectivity, cybersecurity, cloud services, and advanced driver assistance systems within unified software environments, ultimately supporting diverse vehicle requirements while focusing engineering resources on customer-differentiated innovations.

The automotive industry is accelerating its transition toward Software-Defined Vehicles (SDVs), requiring evolution from traditional distributed ECU architectures to domain, zonal, and centralized computing systems. At Renesas Tech Day on October 22, 2026, Inventec unveiled an automotive computing platform built on the Renesas R-Car Gen5 SoC, demonstrating advanced capabilities in system architecture, hardware design, and hardware-software integration. Unlike traditional ECU-centric approaches, SDV development emphasizes shared computing platforms, hardware-software decoupling, and continuous software evolution throughout vehicle lifecycles. The consolidated platform must address computing performance, in-vehicle networking, functional safety, cybersecurity, power management, OTA updates, and long-term maintenance. Renesas, Inventec, and P3 collaborated to showcase how ecosystem partnerships accelerate SDV development. According to Aish Dubey, VP of Renesas's HPC SoC Division, their R-Car SoCs and R-Car Open Access (RoX) enable software reuse across platforms, helping OEMs reduce complexity and accelerate time-to-market. This evolution represents a fundamental shift toward flexible, scalable automotive computing platforms that support both continuous hardware and software advancement throughout vehicle lifecycles.

Connected car technology has evolved significantly beyond basic internet connectivity. In 2026, vehicles are becoming intelligent platforms capable of sensing, communicating, learning, and delivering digital services throughout their operational lifecycle. After two years of isolated pilot testing, automakers have now transitioned to global deployment of over-the-air updates and monetizable software features. The growth is primarily software-driven, integrating AI, cloud computing, edge computing, 5G, and V2X communication technologies rather than relying on hardware advancement. Connected vehicles operate as networked devices that continuously exchange data with cloud platforms, other vehicles, and infrastructure through sensors, cellular networks, GPS, Wi-Fi, and telematics. This sensor-to-cloud-to-application pathway enables critical capabilities including predictive maintenance, remote diagnostics, fleet management, driver assistance, and personalized experiences. Artificial intelligence has advanced beyond voice commands to encompass predictive personalization and generative capabilities, giving rise to a new category: the AI-defined vehicle (AIDV). This transformation is significant for enterprise leaders, OEM engineering teams, fleet operators, telecom providers, and technology partners, as vehicles transition from fixed mechanical products to continuously evolving software platforms that can be updated and improved throughout their entire lifecycle.

# Professional Summary The document "From Ride Hail to Robotaxi" by Cityfi examines the industry transition from traditional ride-hailing services to autonomous robotaxi operations. The analysis provides context on the evolving autonomous vehicle market, tracing how companies currently operating ride-hailing platforms are pivoting toward fully autonomous vehicle deployment. The article discusses the technological advancements, regulatory frameworks, and market dynamics driving this transformation from driver-dependent services to self-driving taxi fleets. Key considerations include infrastructure requirements, consumer adoption timelines, and competitive positioning among major players in the mobility sector. This transition matters significantly as it represents a fundamental shift in urban transportation economics, potentially reducing operational costs through elimination of driver expenses while raising questions about employment, safety standards, and regulatory oversight. The robotaxi model promises improved efficiency and accessibility in metropolitan areas, though successful implementation depends on resolving technical, legal, and social challenges that currently remain unsolved across most jurisdictions.

The automotive industry is undergoing a fundamental shift toward Software-Defined Vehicles (SDVs), where software increasingly determines vehicle functionality, features and customer experiences rather than hardware alone. Traditional vehicles rely on hardware to define capabilities, whereas SDVs use standardized computing platforms where software coordinates vehicle behavior throughout the vehicle lifecycle. This represents a transition from mechanical systems and independent electronic control units to high-performance computing architectures emphasizing connectivity and continuous evolution. Industry standards like AUTOSAR and Eclipse SDV emphasize hardware-software decoupling and modular, extensible software platforms for both in-vehicle and external systems. The critical architectural concept is decoupling, enabling vehicle applications to depend on standardized services and interfaces rather than specific hardware components. While brakes, sensors, motors and structural elements remain essential, their value increasingly derives from software coordination. Importantly, Start of Production (SOP) no longer marks the vehicle's end but begins a continuous software journey. This transformation enables manufacturers to deliver differentiated customer experiences, implement over-the-air updates and evolve vehicle capabilities throughout their lifecycle, fundamentally changing how the automotive industry develops and maintains vehicles.