Electronics Era: ADAS adoption is accelerating globally. How do you see the Indian market evolving, and what factors will drive wider adoption in the coming years?
Sameer Sankhe: India is at an interesting inflection point with ADAS. We are not going to see overnight adoption the way some markets have. The adoption here will be driven by a few very specific forces.
The first is regulation. The government has already mandated a version ADAS for heavy commercial vehicles, covering categories like buses and trucks, requiring systems like Automatic Emergency Braking, Lane Departure Warning, and Driver Drowsiness Detection. That is a real regulatory push, not a theoretical one. When you mandate safety systems for vehicles that cover the longest highway corridors in the country, you create an infrastructure requirement downstream.
The second is consumer awareness. Bharat NCAP has done something important by giving Indian buyers a domestic safety benchmark. When Indian car scores five stars with Level 2 ADAS as standard, it tells the market that these features are moving from premium to mainstream. That shift matters more than any industry forecast.
The third, and this is where I think India’s story gets genuinely different, is the road environment itself. Most ADAS systems globally were designed for structured traffic, consistent lane markings, clear signage. Indian conditions are different. A peer-reviewed SAE study published earlier this year formally documented this divergence. So the question is not just “will India adopt ADAS?” but “will ADAS be adapted for India?” Those are two very different conversations, and the second one is where the real work lies.
Cost engineering will also play a role. India’s automotive industry has always been strong at delivering technology at price points that work for a mass market. That capability will accelerate adoption once the foundational elements, particularly the map and data infrastructure, are in place.
Electronics Era: HD maps are increasingly being recognised as a critical component of ADAS. What role do they play in enhancing vehicle safety and driving intelligence?
Sameer Sankhe: Think of it this way. A camera on a vehicle can see what is in front of it right now. An HD map tells the vehicle what is around the next curve before it gets there. That distinction matters a great deal when you are talking about safety systems that need to make decisions in milliseconds.
HD maps provide the geometry of the road with high accuracy: lane width, lane boundaries, curvature, gradient, road edges, speed limits, traffic sign locations. This is information that sensors alone cannot always capture reliably, especially in conditions like heavy rain, fog, or on roads where lane markings are faded or missing entirely.
Take a practical example. Lane Keeping Assist needs to know where the lane boundaries are. On a well-maintained European highway, the camera can read those markings clearly. On a stretch of Indian national highway where sometimes the paint has worn off, the camera has very little to work with. An HD map provides a virtual lane model, a mapped centreline and drivable corridor, that allows the system to function even when the physical markings are absent.
Similarly, Intelligent Speed Assistance relies on accurate speed limit data. A camera-based system might misread a sign on an adjacent road or miss a sign that is obscured. The map provides a verified, ground-validated speed reference that complements what the camera sees.
Beyond individual features, HD maps enable what you might call predictive driving. The vehicle knows there is a sharp curve ahead, or a steep gradient, or a highway exit, and it can adjust speed or alert the driver well in advance. That is a fundamentally different safety proposition compared to purely reactive, sensor-only systems.
The map is not replacing the sensor. It is giving the sensor context. And in driving, context is what separates a safe response from a late one.
Electronics Era: India’s road environment presents unique challenges compared to global markets. How are mapping technologies being tailored to support ADAS deployment in such conditions?
Sameer Sankhe: This is really the central question for anyone working on ADAS in India, and I think it deserves a candid answer.
India’s road environment is not an edge case. It is a fundamentally different operating context. The ISO 21448 standard, which deals with the safety of the intended functionality of automated systems, treats worn lane markings and construction zones as exceptions. In India, these are the median condition across thousands of kilometres of national highway. India’s Comptroller and Auditor General confirmed in a 2025 report that missing markings and inadequate signage on national highway stretches contribute directly to accident rates. This is not anecdotal. It is audited.
So the mapping technology cannot simply be a global dataset adapted for India. It has to be built from the Indian road surface itself. At Genesys, our approach is ground-truth survey. We use LiDAR, 360-degree street imagery, and a constellation of sensors to capture road features at centimetre-level accuracy. Our HD maps include wide range of road attributes: lane boundaries, curvature, gradient, road edges, traffic signs, speed limits, shoulders, dividers, and more.
What makes this India-specific is not just the data capture but the feature set. We map the drivable corridor, not just the painted lane. We capture road edges where lane markings are absent. We provide mapped centrelines so that a vehicle can hold its position on roads where there is no paint to follow.
We have also aligned our survey framework with the Survey of India’s CORS network, which gives us survey-grade positioning accuracy tied to the national geodetic frame. This matters because when you are building maps that vehicles will rely on for safety decisions, drift and positional error are not acceptable.
The short answer is: you cannot make ADAS work reliably in India by importing an ADAS system developed in the other countries. You need to survey Indian roads and build the map from the ground up. That is what we have been doing.
Electronics Era: Genesys has been at the forefront of developing ADAS-ready maps in India. What are the key technological advancements that have enabled this capability at scale?
Sameer Sankhe: Scale has been the operative challenge. Building an accurate HD map for a test corridor is one thing. Doing it across the national highway network is a different problem entirely.
The first enabler has been our sensor infrastructure. We operate the Genesys Constellation, which is one of the largest collections of mapping sensors in Asia. This includes vehicle-mounted LiDAR systems, panoramic cameras, aerial LiDAR platforms, UAVs, and backpack-mounted sensors for areas vehicles cannot access. Having this range of hardware in-house means we can choose the right capture method for the right environment, which is critical in a country where terrain varies from coastal highways to mountain passes to dense urban corridors.
The second is our intellectual property for using AI and automation in the processing pipeline. Raw sensor data, whether it is a LiDAR point cloud or a 360-degree image, needs to be extracted, classified, and attributed before it can be used to create a usable map. We have invested significantly in AI and machine learning models that extract road features from this data. These models are trained specifically on Indian road conditions, so they handle scenarios that global models typically miss.
The third, and this is increasingly important, is our update mechanism. Roads change. Signage changes. Construction happens. We have developed an Edge-Based Incremental Map Update System that delivers over-the-air delta updates to vehicles, so you are not re-downloading the entire map every time a speed limit changes on a stretch of highway. Different map layers update at different frequencies: dynamic layers like incidents can update in near real-time, while static layers like lane geometry update on a quarterly cycle.
We have invested hundreds of crores to build out our India map product, including sensor infrastructure and the team. It is a significant commitment, but when you are building safety-grade infrastructure, there is no shortcut.
Electronics Era: As automakers transition towards Software-Defined Vehicles (SDVs), how do you see the relationship between mapping, vehicle software, and driver assistance systems evolving?
Sameer Sankhe: The shift to Software-Defined Vehicles changes the role of the map quite fundamentally. In the older model, the map was a static file loaded onto the infotainment system. It told you how to get from A to B. That was essentially it.
In an SDV, the map becomes a living data layer that multiple vehicle systems consume continuously. The ADAS module reads lane geometry and curvature for lane keeping and predictive speed control. The powertrain management system reads gradient data for efficient energy use, which is especially relevant for EVs where range anxiety is real. The navigation system uses the map for routing, but now with AI-driven conversational interfaces that understand natural language, so a driver can say something like “find a good restaurant on the way” and the system handles it intelligently.
We have been building toward this with our Agentic AI Navigation Platform. The idea is that the map is not just a data set but an intelligence layer. It receives inputs, whether from a voice command, from vehicle sensors, or from real-time traffic, and it makes decisions. Route optimisation, charge-aware routing for EVs, predictive alerts, all of this sits on top of the map.
What is also happening, and this is still in its early stages, is the potential for a feedback loop. Vehicles equipped with cameras and sensors are effectively sensing the road as they drive. That sensor data, anonymised and aggregated, can feed back into the map to flag changes: a new construction zone, a changed speed limit, a missing sign. This creates a system where the map and the vehicle improve each other over time.
For OEMs, the question is increasingly about which partners can provide not just data but a platform that integrates with their software stack. The map provider’s role is evolving from a data supplier to something closer to an infrastructure partner. And I think that is a healthy evolution for the industry.
Electronics Era: What are the biggest challenges in maintaining highly accurate and continuously updated maps for ADAS applications, and how can the industry address them?
Sameer Sankhe: This is probably the hardest operational problem in our industry, and it is worth being honest about the scale of the challenge.
India has one of the largest road networks in the world, and it changes constantly. Highways are being expanded, resurfaced, re-aligned. Urban roads see re-channelisation, new flyovers, altered one-way systems. Signage changes. Construction diversions appear and disappear. Keeping a map current across this scale is not a solved problem anywhere in the world, and in India the rate of change makes it even harder.
Our approach is layered. The foundation is survey-grade capture using our sensor constellation. From there, we run AI-driven change detection that compares new imagery and sensor data against the existing map to flag discrepancies. These flagged changes go through validation before they are published into the map.
We also structure the map in layers with different update frequencies. Dynamic information like incidents or temporary closures can update in near real-time. Semi-static information like speed limits and signs updates weekly to monthly. Static geometry like lane models updates on a monthly to quarterly cycle. This layered approach means you are not trying to refresh everything at the same cadence, which is neither practical nor necessary.
The industry-level solution, and I think this is where we are headed, involves collaboration. Vehicles with ADAS cameras are already sensing the road. If OEMs enable that sensor data to flow back, even in aggregated, anonymised form, it creates a powerful closed-loop. The map informs the vehicle, and the vehicle informs the map. We have designed our update pipeline to incorporate this kind of OEM sensor feedback.
But I want to be realistic. We are not there yet in terms of industry-wide data sharing. The frameworks are being built, the conversations are happening, but the practice is still maturing. In the meantime, disciplined survey cycles and AI-driven monitoring are the practical answer.
Electronics Era: Beyond navigation, what emerging use cases do you foresee for HD maps in connected, autonomous, and intelligent mobility ecosystems?
Sameer Sankhe: The interesting thing about HD maps is that once you have captured the physical world at high accuracy, the use cases extend well beyond what people traditionally associate with maps.
The most immediate one is Intelligent Speed Assistance. Over speeding is linked to roughly 68% of road fatalities in India. A reliable, map-based ISA system that knows the speed limit on every road segment and can alert or assist the driver is probably the highest-impact safety intervention available today. The European Union has already mandated ISA on all new vehicles. India may follow a similar trajectory.
Then there is EV routing. This is not just “find me a charger.” It is charge-aware routing that factors in elevation, gradient, traffic, and driving style to predict range accurately and plan charging stops intelligently. Our maps carry the elevation and gradient data that makes this possible.
Urban planning and disaster management is another significant area. We are building 3D digital twins of Indian cities, and these are being used for flood modelling, emergency routing, infrastructure planning, and asset management. The BMC in Mumbai, for example, awarded us a Rs 155 crore contract for a comprehensive 3D model and map stack. Pune, Varanasi, and Ayodhya have also adopted our digital twin platforms for various urban governance applications.
Last-mile logistics is improving with better address resolution. India has had a persistent challenge with address ambiguity, where the same location can be described in multiple ways. With DIGIPIN integration into our map stack, every address gets a unique, geocoded reference, which directly improves delivery success rates and emergency response times.
Usage-based insurance is another area where map data plays a role, by providing context to driving behaviour analytics.
I think the broader pattern is that accurate geospatial data is becoming foundational infrastructure for multiple industries, not just automotive. The map is the common layer.
Electronics Era: How can geospatial intelligence and real-time mapping contribute to improving road safety outcomes in India?
Sameer Sankhe: India loses approximately 1.7 lakh lives on its roads every year. The Union Minister for Road Transport has pointed out that we lose an estimated 3% of GDP to road accidents. These are not abstract numbers. Behind every statistic is a family affected.
Geospatial intelligence contributes to road safety at multiple levels. The most direct one is through ADAS. When a vehicle has access to an accurate HD map, its safety systems work better. Lane keeping, emergency braking, speed compliance, curve warnings, all of these are more reliable when the system has map-based context in addition to sensor input.
But there is also a systemic level. When you have mapped the road network at high accuracy, you can identify where the infrastructure itself is contributing to risk. Our data, combined with findings like the CAG audit that documented missing markings and signage on national highways, can help highway authorities prioritise where to invest in road improvements. You can overlay accident data with road geometry data and identify patterns: stretches where curvature, gradient, or visibility create recurring hazards.
Speed management is probably the single highest-impact area. With 68% of crashes linked to over speeding, a map-based speed alert system that works reliably across the national highway network would have a measurable effect on fatality numbers. This does not require full autonomy or expensive hardware. It requires an accurate map with verified speed limits, integrated into vehicles that are already being sold with basic ADAS features.
Real-time mapping also helps in emergency response. When you know the road network precisely, including lane-level detail, you can route emergency vehicles more effectively. Our digital twin platforms are already being used for this in a few cities.
The honest answer is that no single technology will solve India’s road safety crisis. But accurate geospatial data is one of the more practical tools available, and it is deployable today, not five years from now.
Electronics Era: With increasing collaboration between automotive OEMs, mapping providers, and technology companies, what does the future mobility ecosystem in India look like?
Sameer Sankhe: The mobility ecosystem is becoming more collaborative, and I think that is a positive development. No single company can deliver the full stack alone. You need hardware manufacturers, software platforms, map providers, sensor companies, cloud infrastructure, and increasingly, AI capability, all working together.
What we are seeing in India is the formation of these partnerships. Global ADAS technology providers are entering India through local engineering and manufacturing partners. We have seen several such deals over the past year. Our partnership with NNG, whose automotive software runs in over 60 million vehicles globally across brands like Ferrari, Lamborghini, Mercedes, and Kia, brings a proven in-car software experience to the Indian market combined with our India-grade map data.
I think what is emerging is a model where OEMs focus on their core strengths, vehicle design, manufacturing, brand, while relying on specialised ecosystem partners for data and software layers. The map provider’s role in this ecosystem is becoming more central because the map is the common reference that multiple systems share: ADAS, navigation, EV routing, fleet management, even insurance.
For India specifically, I think we will see the ecosystem develop its own character. The cost structures, the road conditions, the regulatory environment, the consumer expectations, all of these are distinct. The companies that succeed will be the ones that build for Indian conditions rather than simply porting global solutions.
One thing I would add is that this ecosystem also needs to include government and public infrastructure. Regulations like the commercial vehicle ADAS mandate, standards like Bharat NCAP, and digital infrastructure like DIGIPIN all create the framework within which the private ecosystem operates. That public-private interplay will shape the pace and direction of the transition.
Electronics Era: Looking ahead five years, what key trends do you expect will shape the next phase of ADAS, connected mobility, and digital mapping in India?
Sameer Sankhe: I will share what I think is most likely, rather than what sounds most exciting, because I think that is more useful.
First, regulatory momentum will continue. The commercial vehicle ADAS mandate is just the beginning. I expect we will see similar requirements extend to passenger vehicles over the next few years. Bharat NCAP will continue to raise the bar on safety ratings, and ADAS features will increasingly become differentiators in the mass market, not just premium segments.
Second, maps will move from static to continuously updated. The idea of a map that was captured once and loaded onto a vehicle will be replaced by living maps that update over the air, with different layers refreshing at different frequencies. This is already the direction we are moving in, and within five years it should be standard practice.
Third, AI will be more deeply embedded in both the map creation process and the in-vehicle experience. On the creation side, AI-driven feature extraction and change detection will make map updates faster and more efficient. On the vehicle side, conversational AI interfaces will change how drivers interact with navigation and vehicle systems. We are already building this with our Agentic AI Navigation Platform, where the map is not just a dataset but an intelligent layer that responds to natural language, predicts needs, and makes autonomous routing decisions.
Fourth, the EV transition will create new demands on mapping. Charge-aware routing that accounts for real terrain, not flat-road estimates, will be essential. Gradient and elevation data in maps will directly affect range prediction accuracy and driver confidence.
Fifth, and this is perhaps the most significant structural trend, India’s urbanisation will accelerate. The country’s urban population is projected to grow by roughly 300 million over the next 25 years. That scale of urbanisation will require precise geospatial data for everything from road infrastructure to utility planning to emergency management. The companies investing in this data infrastructure today will be the ones best positioned to serve that demand. The building blocks, better ADAS, better maps, better vehicle software, regulatory frameworks, are coming together in a way that was not true even three years ago. The progress will be incremental, but it will be real.
