The question is not if, but when fully autonomous driving will arrive on public roads. The latest Teslas can already do it, and Automated Guided Vehicles (AGVs) are commonplace. The vehicles of the future will combine advanced technologies. Here, you can read about which sensor technologies these are and what their advantages and disadvantages are.
A pilotless airplane or a driverless bus is possible in the foreseeable future. Only legal and psychological objections stand in our way; just as the steam locomotive caused controversy and challenges in the 19th century.
“Cameras and various types of sensors in fused sensor applications are the eyes and ears of the future drivers of our cars,” predicts business development manager Marco Leeggangers.
The evolution of autonomous movement
Autonomous driving was one of the main themes at the IAA Frankfurt this year. The automotive industry is working on technologies that enable completely autonomous movement in public spaces.
The automotive world uses a Scale level from 0 to 5. Level 5 for a fully automated car ride, while you read a book or watch a movie.
According to Leeggangers, all new car models must be automated at level 2 from 2018 onwards to receive a 4- or 5-star safety rating. “The car will then be equipped with Advanced Driver Assistance Systems (ADAS). Such as Automatic Emergency Breaking, Lane Assistance, and Road Edge Detection.”
Tesla has made the leap from ADAS to autonomous in its latest models. The latest version of Tesla's Autopilot is already balancing on the border of level 4 and 5.
Business applications: AGVs
Businesses have long been using autonomously guided vehicles (AGVs) for distribution applications in particular. In many distribution centers, automatic forklifts operate, and order picking is done by robots.
The Netherlands leads in innovation in Agricultural and horticultural automation met UAV's (drones) and AGV's (robots for cleaning stables, feeding livestock, and performing logistical tasks in greenhouses).
Why do we want self-driving vehicles?
Idlers: “In my eyes, this is a logical consequence of technological evolution. Actually, autonomous driving fits with the digital revolution because large amounts of sensor data need to be processed to react independently to the environment. Moreover, the self-driving car is part of the Internet of Things (IoT).”
The benefits of autonomous vehicles are numerous:
- Positive impact on traffic safety. Advanced computers can perform human tasks more efficiently, better, and safer.
- Better utilization of road capacity. Self-driving vehicles drive at shorter distances from each other. This allows them to utilize road capacity more efficiently, reducing and even preventing traffic jams.
- Improved car-sharing opportunities. The use of the self-driving car can be planned so that we can share it. The car for commuting can be available for someone else during the day. Autonomous driving will boost the predicted sharing economy.
- Sustainability: AGVs perform their tasks more efficiently than humans and save raw materials and energy in various industries.
- Productivity: An AGV never gets tired, can handle heavier tasks, and operates flawlessly.
- Cost savings: AGVs enable the full automation of distribution processes. Mobile robots also help reduce costs in agriculture and horticulture.
Detection challenges for distance measurement and positioning
To enable a vehicle to drive autonomously, it needs a comprehensive view of its surroundings. There are four detection challenges for dynamically generating an environmental model.
- 1. Determining the clear passing space on the road surface.
- Determining the geographical route via the navigable space.
- 3. Detecting moving objects (other road users and moving obstacles).
- 4. Recognizing and interpreting road signage, such as traffic signs, traffic lights, road markings, and other visual cues.
Sensor technology has advanced so much nowadays that there are solutions for all detection challenges.
Detection tools for autonomous vehicles
For autonomous driving and advanced driver-assistance systems, primarily radar, lidar, and sonar sensors applied. Combined with cameras and GPS, a vehicle thus dynamically scans its environment. Smart software processes the large amount of data, allowing it to always know its position relative to objects.
These techniques are possible because processors have become increasingly powerful and smaller.
Sensor technology development
Leeggangers indicates that Sentech plays a role in the development and R&D of sensor technology for AGVs. “For example, we already use radar, lidar, and ultrasonics in distance sensors and orientation sensors. As an independent sensor integrator, we are now working on integrating radar and lidar into compact ‘fused’ sensor applications.”
According to the Business Development Manager, sensor fusion leads to smarter and better customer applications, specifically in the area of autonomous movement.
Pros and cons of sensor techniques
The most promising sensor technologies for self-driving vehicles are lidar and radar. Lidar scans the environment with light (laser or infrared), while radar does so with radio waves. “The development of lidar and radar is progressing very rapidly. This is because processor chips are getting smaller and the technology has become more affordable,” according to Leeggangers.
Lidar has significant advantages in remote sensing. One of these is its high resolution, which is necessary for accurately detecting stationary and moving objects. On the other hand, weather conditions like fog and rain have a greater negative impact on accuracy. “Lidar is suitable for observing moving objects in the immediate vicinity of a vehicle,” explains Leeggangers.
Radar can see further, but as the distance increases, accuracy decreases. Therefore, according to him, radar is more suitable for remotely detecting moving objects in front of the vehicle.
The future of self-driving vehicles
“What's special is that the technological visions of car manufacturers differ. One prefers lidar, another prefers radar. The car manufacturers have a sensor-based system as a common starting point. We see a future with advanced fusion sensors in integrated sensor applications,” says Leeggangers.
He also sees new players on the autonomous driving market with a different technological approach, such as Google and Intel. Google has developed its own 3D technology, based on route information and 3D maps.
Intel, the processor manufacturer, has entered the autonomous driving market with the acquisition of Mobileye. The technology concern expects its first self-driving car on public roads in 2021. Intel uses the most advanced visual technology (cameras and software) in vehicles for environmental perception.
However, Leeggangers expects sensors to remain critical links in autonomous driving technology. “You will always need redundant sensor systems to supplement camera or GPS systems. No matter how advanced, anything can break. Redundancy will therefore become increasingly important as the fleet evolves toward full autonomy and driverless traffic.”
More about the development of lidar and radar
Sentech is focusing heavily on the further development of lidar and radar sensors, with an emphasis on sensor fusion. These are the most suitable sensor solutions for autonomous movement in public spaces and business environments.
Sensor fusion is the ultimate form of integration and enables next-generation automotive applications.
