Access Official Technical PDF

Share Summary:LinkedInEmail

Israeli automotive technology companies featured in this industry overview work across a broad set of vehicle systems, from far-infrared cameras and LiDAR to computer vision, radar, driver monitoring and on-demand transit software. Their approaches address a central engineering challenge: helping vehicles interpret their surroundings and occupants, then turning that information into useful alerts, decisions or services.

For automation engineers and technical decision-makers, the technologies also illustrate how a complex system can combine sensors, algorithms, embedded computing and user interfaces. The examples range from single-purpose detection tools to platforms that bring together several data sources. They are not interchangeable solutions; each addresses a different part of the vehicle’s operating environment.

Israeli automotive technology for vehicle perception

Vehicle perception systems convert sensor input into information about roads, objects and other conditions. The companies in this overview use several sensing methods, each with its own role. AdaSky, for example, describes a far-infrared camera paired with computer vision algorithms. Its system adds information from a different part of the electromagnetic spectrum and is designed to detect and classify surrounding objects at near and far range, including pedestrians and animals in varied driving conditions.

Other approaches use multiple cameras or active sensing. Corephotonics’ TransportEye is a forward-facing stereoscopic camera system intended to provide depth information for driver assistance and autonomous driving applications. The company describes a compact form factor designed to fit behind the windshield and rearview mirror. Innoviz, meanwhile, develops solid-state LiDAR products for automotive applications, with the stated goal of supporting broader adoption through product size and cost considerations.

RFISee focuses on radar-based active driver assistance. Its described arrangement uses six to eight connected sensors around a vehicle to build a 3D, 360-degree view of the surroundings. The system is intended not only to alert drivers to hazards but also to interact with vehicle safety systems. These examples show why sensor selection is an architectural decision: the desired information, installation constraints and downstream functions all affect which sensing approach is suitable.

Combining sensors and algorithms

A sensor alone does not provide a complete interpretation of a driving scene. Software must process the incoming data, recognize relevant objects or conditions, and make that output available to other vehicle functions. Vayavision’s platform illustrates a sensor-fusion approach, using raw samples from cameras, LiDAR and radar to construct a 3D model of the environment. AI algorithms then operate on that model to provide an environmental representation.

Sensor fusion is also relevant when a system combines visual and non-visual information. MobiWize describes dynamic mapping for advanced driver assistance and autonomous driving. Its solution uses local and crowd-sourced sensor data to model road characteristics, complementing vision-based sensors with other inputs. The resulting embedded maps are intended to provide real-time insights into road conditions and vehicle performance.

Software efficiency is another design consideration, particularly when perception must run on vehicle computing hardware. Brodmann17 offers software-only deep-learning algorithms for computer vision applications such as people and vehicle detection and behavior analytics. The company emphasizes reducing computing requirements while maintaining recognition accuracy. IonTerra similarly describes real-time visual analysis on low-computational-resource devices, with on-the-fly reprogramming to support detection of additional objects and location- or time-dependent tasks.

Inside the cabin: monitoring and interaction

Vehicle systems must account for the driver and passengers as well as the external environment. EyeSight’s computer vision and deep-learning solutions address driver awareness, infotainment control and driver monitoring. The described functions include detecting drowsiness or inattentiveness and adjusting the in-car environment according to detected preferences and needs. The processing is designed to run locally and in real time.

Jungo’s CoDriver also uses computer vision and deep learning to monitor drivers, passengers and the vehicle cabin. Neteera takes a different sensing approach: its sub-terahertz technology is described as a contact-free method for measuring vital signs, including heart rate and respiration, from a distance through clothing and furniture. The company identifies automotive as one of its application areas.

Voice interfaces form another part of the cabin system. VocalZoom develops human-machine communication sensors for voice authentication and voice control in noisy environments. Potential uses described by the company include interacting with in-car virtual assistants, authenticating a passenger and enabling voice-based services. These systems make interaction a technical integration issue as well as an interface choice: reliable input depends on capturing and interpreting signals in the vehicle environment.

From autonomous driving to transportation services

Some platforms target the full driving stack rather than a single perception function. Imagry describes a cameras-only self-driving platform based on deep inverse reinforcement learning. Its approach is intended to learn from real-world and simulated data without manual labeling, extracting driving parameters and an abstract model of the road to train the vehicle’s driving model. The company positions the system for complex, previously unseen scenarios.

Other technologies extend beyond the vehicle itself. Via’s web and mobile platform is designed for on-demand shared transportation. It matches ride requests with vehicles traveling in the same direction and provides tools for transportation managers, including vehicle-utilization and driver-workflow functions. This represents a different layer of mobility technology: coordinating passengers and vehicles rather than interpreting a driving scene.

Practical considerations for technical teams

Together, these examples highlight several considerations for teams evaluating automotive systems:

  • Define the information required. Object detection, depth estimation, cabin monitoring and road mapping are distinct functions.
  • Plan for system integration. Sensor outputs must connect to perception software, alerts, vehicle controls or service platforms as required by the application.
  • Account for computing constraints. Real-time processing and embedded deployment make algorithm efficiency and hardware demands relevant design factors.
  • Consider operating conditions and installation. Sensor placement, environmental conditions and the intended field of view shape system design.

The practical implication is that advanced automotive functions depend on coordinated choices across sensing, software, computing and human-machine interaction. Israeli automotive technology companies in this overview address different pieces of that system, from individual sensors to perception platforms and transportation services. For engineers and decision-makers, comparing those functions against a defined vehicle requirement is a more useful starting point than treating any single technology as a complete solution.

Access Official Technical PDF

About The Author