Road safety is and continues to be a critical issue in India. Although the government has implemented various road safety improvement programs, the overall situation is far from satisfactory. The country witnesses 415 deaths per day in road accidents – the highest in the world.
Today, vehicle and passenger safety features are given precedence alongside other features like size, mileage, transmission, etc. when buying a car. The automotive industry is increasingly adopting safety systems like Advanced Driver Assistance Systems (ADAS) and Automotive Radar that alert drivers of potential risks and play a preventive role in mitigating accidents.
TechStory connected with Piyali Goswami, Software Development Manager, Texas Instruments (TI) India, to understand how TI is bringing about a change in the automotive safety space, towards helping eliminate human error and mitigate accidents.
“ADAS solutions extract data from a diverse set of sensors and this data is converted to actionable intelligence for the vehicle. The sensors include at least a minimum of different types of cameras and associated optics, radars, and ultrasonic technology,” said Piyali. “Advanced versions also include LiDAR and thermal night vision. The system also performs vehicle localization by comparing features that are extracted from sensors against maps.”
For over a decade, feature-based computer vision algorithms have analyzed images for ADAS operations. Challenges like reduced traction due to wet weather, or other extreme situations, worn pavements, and unsafe road conditions, require vehicles to adapt in real-time. These scenarios cannot be coded and need a dynamic system that can help the car sense, interpret, and react quickly to the world around it. Goswami says that “such systems require data and the ability to process the data in real-time using a combination of computer vision and efficient AI/ deep learning neural networks.”
Semiconductors – the lifeblood of ADAS technology
Semiconductors are revolutionizing the automotive industry as it plays a major role in the electrification and incorporation of autonomous technologies.
“A common misconception going around has people assuming automation means highly or fully self-driving vehicles, which are typically categorized at levels 4 and 5 of autonomy by the Society of Automotive Engineers (SAE),” according to Goswami. “However, we need to understand that there are several intervening levels before we hit self-driving cars. These vehicles will arrive somewhere in the future, but where we are today, the technology is helping cars slowly progress through the SAE levels of autonomy.”
“Preparing for random events is only half the battle when ensuring that the product meets functional safety requirements.” Goswami opines that there has been an emphasis on electrical components performing critical functions from fluid level monitoring in chemical plants to autonomous vehicle navigation and aircraft lift control. The Semiconductor industry plays a key role to enable this. “Semiconductor engineers anticipate potential safety hazards, possible causes of failures at Circuit, System, and Software level, their likelihood of occurrence, and how such risks can be reduced by using integrated circuits (ICs) that support functional safety.”
Evolving vehicular safety standards
Speaking about the various automotive mobility solutions and technologies, Goswami explained that TI is currently developing technologies that enable ‘Smarter cars and fewer collisions’. This ADAS technology helps cars take corrective actions just like a driver would, by sensing weather conditions, detecting and understanding the objects on the road (people, traffic posts, electric posts, etc.), and making real-time decisions. Such features include automatic emergency braking, driver monitoring, forward collision warning, and adaptive cruise control.
“We are also working on Radar on a chip. Radar technology has historically been large, expensive, power-hungry, and designed to detect only large objects like planes or tanks – not pedestrians”, Goswami said. “We have developed a high-frequency radar sensor that is of the size of a coin. It can not only detect an obstacle but can also differentiate between objects such as a car and a pedestrian or a ball and an animal sleeping on the road, etc. It also meets the latest automotive safety standards, such as the New Car Assessment Program (NCAP), on a single chip.”
The European New Car Assessment Programme or Euro NCAP, a car safety performance assessment programme founded in 1996. The Euro NCAP provides guidance on development and activities for the automotive industry, as well as other stake holders involved in the ecosystem. The programme’s 2025 roadmap shows the different safety systems that will come into play in the coming years, making the driving experience safer.
She also stated that this radar technology is a good fit for applications such as child presence detection, driver monitoring, and passenger classification (child or adult), and is smarter than the current safety technologies (e.g.: seat-belt reminders and speed alerts) currently in use. Texas Instruments is developing devices that enable Autonomous Emergency Braking (AEB), Driver Monitoring, Automatic Emergency Steering, V2X communication, Enhanced rear end crash protection, Front camera systems for pedestrian and cyclist safety and Radar based child presence detection.
As vehicles grow smarter by connecting, communicating, monitoring, and making decisions that help reduce and prevent accidents, the computing power required to process this data has skyrocketed.
Goswami shares that TI’s Jacinto processors help improve awareness of the car’s surroundings and accelerate data-sharing. This efficiently manages multilevel computing in real-time while reducing power-hungry memory transfers. TI is also working on technology domains that can enable cybersecurity to ensure secure and an operational fail-safe working in vehicles.
Accelerating performance with automation
Although automation can be introduced in any vehicle, self-operating vehicles will require numerous sensing and actuation systems.
“Cars can also be designed to operate automatically with greater fuel efficiency than most human drivers can achieve, especially in a situation where maintaining a constant speed is difficult. I feel HEV/EV development for automakers will be closely linked with the development of advanced driver assistance systems (ADAS) and self-driving capabilities, from overall safety, fuel economy, and emission control perspective.”
Road Safety, though extremely important, has still not been able to attract due attention in India. According to a report by the World Bank, India has 1% of the world’s vehicles but 11% of the global deaths from road accidents. We asked Goswami if ADAS and Driver Monitoring Systems (DMS) can eliminate accidents caused by human error.
“Drivers are the most common failure point in vehicle-related accidents. This has also resulted in a greater volume of research and Euro NCAP safety initiatives to be undertaken so that common driver errors can be eliminated. Our DMS provides a significant level of feedback to ADAS Electronic Control Units (ECUs), and Autonomous Driving Systems to correct common human errors. For instance, if a driver is distracted a vehicle that has our DMS system alerts the driver or maneuvers the car to avoid a collision. Our DMS solutions use multiple cameras equipped with infrared illuminators directed toward the driver, which captures quality images that can be processed even in low-light conditions.”
TI’s contribution to automotive safety is cutting-edge and has the potential to change the landscape completely.
Automotive manufacturers are dependent on semiconductor suppliers like TI to develop innovative technology and solutions to enable the evolution of such functionalities. “Because ADAS technology is changing rapidly, automakers need solutions that provide extra performance and offer the flexibility to integrate new features with minimal design changes. With its extensive experience in creating products for automotive electronics, TI can offer optimized, complete solutions to ADAS developers that will simplify design and reduce the component count.”
