At Auto EV TVS Summit, 2025, a panel of industry leaders—from semiconductor companies to automotive software firms—gathered to discuss one of the most transformative shifts underway in mobility: the rise of the Software-Defined Vehicles (SDVs). Moderated by Mohammed Saeed Mombasawala, CTO at Keysight Technologies, the discussion brought together voices from Bosch Global Software Technologies, Marelli India, NXP Semiconductors, Aumovio, and Auto Ascent to unpack how vehicles are evolving from mechanical machines into continuously upgradeable software platforms.
The consensus across the panel was clear: software is no longer an auxiliary component in the automotive stack—it is becoming the primary architecture around which the vehicle is built.
The Shift Toward “Intelligence on Wheels”
Software-defined vehicles represent a departure from the traditional automotive model, where functionality was fixed at the time of manufacturing. Instead, SDVs rely on software layers that can evolve through updates, new services, and data-driven improvements throughout a vehicle’s lifecycle. “SDV is essentially about delivering affordable intelligence on wheels,” explained Bosch’s Naved Narayan during the panel discussion.
The concept is already beginning to take shape in India. Features such as over-the-air updates, connected vehicle services, and Level-2 driver assistance systems are gradually entering the market. However, unlike Western markets where premium vehicles dominate adoption, India’s automotive ecosystem operates under a different constraint: cost sensitivity.
Industry participants emphasized that the success of SDVs in India will depend on achieving a delicate balance between technological sophistication and affordability.
India’s SDV Journey: Progress, But With Constraints
While the SDV revolution is global, India’s pathway has unique challenges. Panelists noted that the country is progressing toward connected and intelligent mobility, but several structural barriers remain.
Latha Chembrakalam, Founder and CEO of Auto Ascent, highlighted that the industry is driven by three key factors: safety, ease of use, and joy of driving. Software-defined architectures promise to enhance all three—but India must simultaneously contend with infrastructure gaps and regulatory limitations.
“India is a cost-sensitive market, and infrastructure readiness also plays a major role,” she noted. “While progress is visible, technological capabilities and regulatory frameworks still need to evolve to fully support SDV adoption.” Yet there are encouraging signs. Automakers such as Mahindra and MG have already begun introducing advanced connected features and driver assistance technologies in the Indian market, creating early momentum.
The New Architecture of the Vehicle
At the engineering level, SDVs are forcing a fundamental redesign of vehicle electronics. Traditional vehicles rely on dozens of electronic control units (ECUs), each responsible for a specific function. SDVs, however, are shifting toward centralized computing architectures, where domain controllers or zonal controllers manage multiple vehicle functions.
This transformation also introduces new challenges. Automotive companies must reconcile modern SDV architectures with legacy platforms and existing software stacks. “For new electric vehicle platforms, it is easier to design from scratch,” Chembrakalam explained. “But the real complexity lies in integrating SDV capabilities into existing vehicle platforms with legacy systems.”
In addition, the industry still lacks fully standardized architectures across OEMs, Tier-1 suppliers, and technology vendors. Without stronger coordination, experts warned, the ecosystem risks becoming fragmented.
Sensors, Connectivity, and the Edge Computing Challenge
Software-defined vehicles depend heavily on three technological pillars:
- Sensors
- Connectivity
- In-vehicle compute
Radar, cameras, LiDAR, and other sensing technologies collectively form the vehicle’s perception system. Rather than relying on a single sensor type, most companies now favor sensor fusion architectures that combine multiple sensing modalities. Rajkumar Anantharaman of NXP noted that radar remains one of the most critical sensing technologies due to its reliability across weather conditions. However, cameras provide complementary information, such as object classification and visual context.
“Radar alone cannot solve everything,” he said. “The industry is moving toward fusion architectures where radar and camera data are combined to improve environmental perception.” At the same time, sensing systems are also becoming increasingly intelligent. Modern radar chips now integrate edge processing capabilities, allowing them to detect and classify objects directly on the sensor rather than transmitting raw data to a central processor.
This shift toward edge AI processing helps reduce latency and bandwidth requirements.
Connectivity: Why Latency Matters
Connectivity plays another crucial role in the SDV ecosystem. While current vehicle platforms rely primarily on 4G networks, panelists believe 5G—and eventually 6G—will enable new levels of vehicle intelligence, particularly for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2X) communication.
The reason is latency. Even a delay of a few hundred milliseconds can significantly affect vehicle safety systems. In collision scenarios, a 200-millisecond delay could translate into several meters of additional braking distance. Future networks promise to reduce latency to just a few milliseconds, enabling faster information exchange between vehicles and surrounding infrastructure.
However, experts cautioned that India’s current telecommunications infrastructure still needs to mature before such capabilities can be widely deployed.
AI, Data, and the Digital Twin Revolution
Perhaps the most complex dimension of SDVs lies in software development itself. Unlike traditional automotive software, SDV platforms require continuous integration, machine learning models, and large-scale data pipelines.
Training autonomous and driver-assistance systems requires massive datasets capturing real-world driving conditions. Yet India presents a unique challenge in this area: driving environments that are far less structured than those in Western markets. Unpredictable traffic behavior, unmarked roads, and unusual scenarios—from animals crossing highways to dense urban congestion—make it difficult to collect sufficient real-world training data.
To overcome this limitation, companies are increasingly relying on digital twins, simulation environments, and synthetic data generation. These virtual environments allow engineers to simulate thousands of driving scenarios before deploying software in actual vehicles. By shifting validation earlier in the development process—known as “shift-left engineering”—companies can test and refine software models without relying entirely on expensive physical vehicle testing.
Vehicles That Improve Over Time
One of the most intriguing aspects of SDVs is the possibility that vehicles may increase in value over time. Traditionally, a car’s capabilities remained fixed after leaving the factory. With SDVs, however, software updates can introduce entirely new features years after purchase.
Bosch’s Narayan described this shift as “an upgrade without actually upgrading the vehicle.” Through over-the-air updates, automakers can introduce new driver assistance features, improved algorithms, or additional digital services long after the vehicle has been sold.
At the same time, consumer expectations are rapidly evolving. According to Yogesh Davangere Adevappa, rising awareness of global technology trends—driven by social media and digital exposure—is pushing buyers to expect more intelligent and feature-rich vehicles. “People are increasingly aware of the technologies available worldwide,” he noted during the panel discussion. “That awareness is driving demand for vehicles with more connected features, safety systems, and intelligent capabilities.”
Why the Software-Defined Vehicle Matters
Ultimately, the SDV transformation is about more than just technology. It represents a fundamental redefinition of what a vehicle is. For consumers, the appeal lies in enhanced safety, convenience, and personalized driving experiences. For automakers, SDVs open the door to entirely new business models built around software services and continuous updates.
There is also growing demand for customization in vehicles. Vikram Bhatt, Aumovio, pointed out that SDV architectures allow drivers to configure vehicle behavior dynamically—whether enabling features like park-assist modes, obstacle detection systems, or personalized driving configurations. These runtime adjustments represent a fundamental shift from static vehicle functions to software-enabled experiences.
Yet despite this progress, panelists acknowledged that India is still at an early stage of SDV deployment compared to global markets.
As one panelist summarized during the discussion, the future of mobility may not just be electric or autonomous—it will be software-driven at its core.
