Assessing the latest developments in autonomous transport technologies

Autonomous transport technologies moved decisively from concept to implementation at CES 2026, as vehicle manufacturers, technology suppliers and mobility providers showcased systems already operating in real-world environments

Announcements from Karsan, Texas Instruments (TI) and the Volkswagen Group illustrated how autonomy is advancing simultaneously at the vehicle, semiconductor and data ecosystem levels, with a shared emphasis on safety, scalability and regulatory readiness.

FIELD-PROVEN AUTONOMOUS PUBLIC TRANSPORT

Public transport specialist Karsan used CES 2026 to underline that autonomous mobility is no longer confined to pilot projects. The company presented next-generation autonomous vehicles already operating in live service and formally introduced its Karsan AI (Autonomous Intelligence) vision.

Defining this approach, Karsan CEO Okan Baş said: “Karsan AI is positioned as a mobility intelligence that perceives, makes instant decisions and continuously learns. The future is not only electric, but also intelligent. This approach reveals Karsan’s vision of not only producing zero-emission vehicles, but also designing the future of cities through autonomous and intelligent transportation solutions.”

Karsan positioned its autonomy strategy as a natural evolution from electrification to smart mobility, emphasising end-to-end system integration and long-term partnerships. According to Baş, “The secret to Karsan’s autonomous success lies in strong partnerships and an end-to-end solution approach.”

Operational pressures are accelerating adoption. “The difficulty of finding drivers and increasing operational costs worldwide are pushing cities towards autonomous solutions,” Baş said. He added that Karsan’s focus remains on regulation-aligned deployment: “We are managing this transformation in a manner that is compliant with regulations, safe and sustainable. Our goal is to implement truly driverless public transport systems where the safety driver is completely eliminated, in parallel with the development of regulations.”

Karsan used CES to reinforce its broader message that autonomy must deliver tangible benefits. “The mobility of the future will not only be electric; it will also be intelligent, autonomous and add real value to human life,” Baş said.

SEMICONDUCTORS ENABLE HIGHER LEVELS OF AUTONOMY

At the component level, Texas Instruments highlighted how advances in automotive semiconductors are enabling more capable autonomous systems across a wider range of vehicles. TI introduced new high-performance computing system-on-chip (SoC) devices, radar sensors and in-vehicle networking technologies designed to support up to SAE Level 3 autonomy.

“The automotive industry is moving toward a future where driving doesn’t require hands on the wheel,” said Mark Ng, director of automotive systems at TI. “Semiconductors are at the heart of bringing this vision of safer, smarter and more autonomous driving experiences to every vehicle.”

Central to this strategy is TI’s scalable TDA5 SoC family, delivering AI performance from 10 to 1200 TOPS with power efficiency beyond 24 TOPS/W. The devices integrate the latest generation C7 neural processing unit and Arm Cortex-A720AE cores, supporting sensor fusion, AI decision-making and cross-domain integration within a single chip. According to TI, this architecture reduces system complexity while helping automakers meet Automotive Safety Integrity Level D requirements.

Radar capability was another focus, with the launch of the AWR2188 eight-by-eight 4D imaging radar transceiver. By integrating eight transmitters and eight receivers on a single chip, the device simplifies high-resolution radar design while improving detection performance. TI stated that the transceiver enables object detection beyond 350 metres and supports advanced use cases such as identifying closely spaced vehicles and detecting lost cargo.

Vehicle networking is also evolving to support autonomy. TI’s DP83TD555J-Q1 10BASE-T1S Ethernet PHY extends Ethernet to vehicle edge nodes, enabling unified, time-synchronised data exchange with reduced wiring complexity. Together, TI described these technologies as an end-to-end system approach supporting higher automation levels across entry-level to premium vehicles.

LEARNING FROM REAL-WORLD DRIVING DATA

While hardware and vehicle platforms advance, data-driven learning is becoming increasingly central to autonomous system performance. The Volkswagen Group announced plans to expand its use of anonymised sensor and image data from customer vehicles across around 40 European countries starting in January 2026.

The objective is to continuously optimise driver assistance systems and automated driving functions using real traffic scenarios rather than simulations alone. According to the Group, customers will benefit through software updates that enhance comfort and safety, with customer consent described as a “fundamental prerequisite” and data use governed by national and European regulations.

Volkswagen’s existing fleet already contributes to safety improvements through anonymised swarm data used to generate high-resolution maps, enabling lane guidance on roads without markings and refined hazard alerts. Engineers now aim to deepen this approach by analysing specific scenarios where assistance systems are most critical, such as interactions with pedestrians and cyclists or complex parking environments.

Data transmission is event-based rather than continuous and can be triggered by emergency braking, full manual braking or sudden evasive manoeuvres. Relevant inputs include camera images, sensor detection results, vehicle dynamics data and environmental conditions. The Group emphasised that individual identification of people is not relevant and that consent can be revoked at any time.

CONVERGING PATHS TOWARD AUTONOMY

Taken together, the announcements at CES 2026 demonstrate how autonomous transport is progressing through parallel developments in vehicles, electronics and data ecosystems. The technology’s deployable capability is increasing, built on proven technology, regulatory alignment and continuous learning from real traffic conditions.