Advertisement

Sensor technology will have an increasingly important role to play in the vehicle models we’ll be driving in the future. Their number and extent of their variety is continuing to expand all the time. It’s estimated that by 2020, automotive manufacturers will be utilizing more than 10 billion sensors each year.

There are several key forces that currently influence the automotive sector, and each of these will rely heavily on the acquisition of data in an accurate, repeatable, reliable, and timely manner. The most prominent are:

Energy efficiency: This is mainly being driven by international environmental legislation, with fundamental changes in how cars are constructed so that fuel economy metrics can be raised and reductions in vehicle weight can be realized. Conventional mechanically-based systems are thus being usurped by electronic ones. Magnetic sensors with strong linearity must be specified so that the degree to which the throttle is actuated exactly matches the pressure applied to the pedal. Likewise, magnetic sensors with superior angular resolution are needed to determine the position of the steering wheel.

Downsizing of engines: This is another development with serious implications in relation to the sensors located under the hood. To further improve fuel economy, car manufacturers are making use of turbochargers and direct fuel injection to get more power out of smaller combustion volumes. The resulting reduction in engine dimensions causes increased heat levels. Engine management systems expect tighter tolerances at elevated temperature levels, which is setting new challenges for the sensor systems employed.

Migration towards HEV/EV: There’s a growing prevalence of hybrid electric vehicle (HEV) and electric vehicle (EV) models. Forecasts from Strategy Analytics suggest that by 2025, this will constitute 15 percent of all car sales. Other reports predict that between then and the 2040 timeframe, this figure will more than triple. The uptake of HEVs/EVs will also have a significant impact on how automotive operations are performed. The higher current densities present as power inverters become more compact (so that the purchasing price of these vehicles can be made as appealing as possible) mean that current sensors incorporated need to cope with more exacting demands.

Increased concerns about safety: To address this, we’re going to see more sophisticated sensor technologies being installed. Among these will be optical time-of-flight (ToF). ToF-based gesture recognition is already starting to be featured in some higher end car models, but that’s only the beginning. In addition, there’s considerable scope for such technology to be taken and further utilized for driver monitoring purposes. Through this, the car’s advanced driver assistance system (ADAS) will be able to check if the driver’s concentration is actually on the road ahead or whether it’s necessary to intervene. Not only will this lead to significant improvements in safety levels, it’ll bring us closer to the long-term objective of autonomous driving.

Along with the emergence of sensor mechanisms (such as ToF) that haven’t been engaged in an automotive context before, the progression towards autonomous vehicles will also rely on the further adaptation of established automotive sensing technologies. Currently, ultrasonic sensors are mounted into vehicle bumpers to serve in assisted parking. Here they only need to function at very low driving speeds (less than 5 km/hour) and aren’t required to measure the small distances involved with 100 percent precision. Autonomous driving will be far more demanding. Here it could potentially be used in combination with various other sensing technologies to provide accurate distance measurement at higher speeds.

There’s no doubt the arrival of autonomous vehicles has the potential to make our roads much safer, as it will eliminate driver error from the equation (which is by far the biggest cause of car accidents). An autonomous driving system will retain much more information on what key decisions need to be made, allowing quicker reaction to potentially life-threatening situations. Having said that, it may not happen as quickly as the current media hysteria seems to imply.

Autonomous vehicles will need access to a comprehensive array of different sensors, each providing the necessary data that will aid its decision-making process. Much of the relevant sensor technology is already available (or certainly on the horizon), but how the captured data is subsequently processed represents the biggest challenge. Current image processing technology isn’t able to determine the complex environments involved within a short enough time to support true autonomous driving, which makes it likely that LiDAR will play an important role.

The adaptive cruise control systems found in contemporary car models utilize radar to measure the distance to the vehicle in front. This is adequate enough when the vehicle is on the highway but when placed into an urban/suburban environment, the distances are a lot shorter and pedestrians/vehicles can approach from other directions. LiDAR functionality (like radar) is based on measuring the reflection and absorption of a transmitted signal. While radar relies on radio waves, LiDAR uses lasers. The time that elapses between the transmission of a laser pulse and when its reflection is received allows the distance to the object to be calculated. In addition to covering multiple directions, LiDAR can deal with much smaller objects than radar.

Contrary to camera systems (which view the environment in focal planes), LiDAR delivers an accurate detailed 3D profile. This method makes it much easier to distinguish objects in front or behind them, regardless of the lighting conditions at that time. Furthermore, the imaging data can be captured and processed much faster. As LiDAR technology begins to mature and the costs reduce, there will be greater interest in its application within vehicle designs.

Ultimately, there needs to be significant improvements over the course of the coming years in relation to the key performance parameters of sensors like dynamic range, accuracy, responsiveness, etc. At the same time, it’s clear that certain attributes like robustness and ease of integration, will also have heightened value. It’s no longer going to be about just providing the sensor devices, but to deliver complete solutions (whether these are customized or standard of the shelf ones). As tolerances tightens and signal-to-noise ratios become more critical, car manufacturers will require enhanced sensor interface technology to accompany the sensing element. There are also certain to be opportunities for sensor fusion, so that data from multiple sources can be simultaneously analyzed. Through the advances now being made in sensor devices, the automobile industry will bring safer models to market that are more fuel efficient, and offer greater comfort. In time they will also enable the unquestionable benefits of vehicle automation to be derived.

Advertisement
Advertisement