listen to this article
Bicyclists can present unique challenges to human drivers. Their behavior can be difficult to predict and they may change lanes or swerve to avoid obstacles faster than other drivers. In 2020, fatal cyclist accidents increased 5% (846) compared to 2019, according to NHTSA.
Autonomous vehicles face those same challenges, so companies making self-driving systems need to be especially careful when considering cyclists. While self-driving cars have the potential to reduce vehicle accidents and make roads safer, they must first learn the best way to navigate the road.
Argo AI, a company working on its own autonomous driving system, in collaboration with The League of American Bicyclistslaunched his own best practices to create autonomous driving systems that work safely around cyclists.
“Argo AI and the League of American Bicyclists share a common goal of improving street safety for all road users,” said Ken McLeod, policy director for the League of American Bicyclists. “We appreciate Argo’s proactive approach to researching, developing and testing the safety of people outside of vehicles. Roads have become significantly less safe for people outside of vehicles in the last decade, and by addressing interactions with cyclists now, Argo is demonstrating a commitment to the role of automated technology in reversing that deadly trend.”
These are the six suggested guidelines to follow when developing autonomous driving systems.
For an autonomous car to react to a cyclist on the road, it must first understand what cyclists are and how they move. Cyclists don’t behave like anything else on the road, so it’s important for an autonomous driving system to designate them as a central object representation within its perceptual system.
By tagging a diverse set of bike images, an autonomous system can recognize a rider from any point of view, speed, and bike position or orientation. Autonomous systems should also be able to recognize different shapes and sizes of bicycles, such as recumbent bikes or electric bikes.
There are certain behaviors that are common to cyclists that autonomous driving systems should be able to recognize and anticipate. Lane splitting, yielding at stop signs, or bicycling are all behaviors that autonomous driving systems must be able to recognize and react to appropriately.
Autonomous driving systems must use specific forecasting models for cyclists. With these models, when a car encounters a cyclist, it can predict many potential paths for the cyclist, making it easier for the car to predict and respond to the cyclist’s actions.
This video from 2019 shows a Waymo self-driving vehicle approaching two cyclists and a vehicle blocking a bike lane. The autonomous driving system correctly predicts that cyclists will pass the vehicle on the left and slows down to allow them to pass.
Many cities and states have bike-specific laws that a self-driving car must be aware of. For example, in some states bicyclists may treat red lights as stop signs. Additionally, mapping of self-driving cars must include any bike lanes. Knowing where the bike lanes are will allow a self-driving car to anticipate more cyclists in those areas and be more attentive to common bicyclist behaviors.
The goal of autonomous cars is to replicate and improve human driving. This means that autonomous vehicles must communicate with cyclists and other drivers on the road in the same way that a human driver would. Autonomous cars should use turn signals when appropriate to help others understand their intentions.
Autonomous cars must also maintain a greater following distance from cyclists in the same way that a human driver would. They must also be especially careful when passing bicyclists.
It’s inevitable that sometimes drivers and cyclists act unpredictably and self-driving cars aren’t sure how to react. When this happens, the car should slow down and create more distance between the car and the bicyclist, if possible. Autonomous systems must always take uncertainty into account.
Extensive testing is one of the most important things when developing an autonomous driving system. Testing both in simulation and in the real world is crucial to creating a system that can safely operate around cyclists.
Simulation should be used to test real life scenarios in a virtual world to test different scenarios safely. These scenarios must capture the behavior of vehicles and cyclists, as well as changes in road structures and visibility. Real-world tests should be used to validate simulations and ensure that the technology behaves in the same way as it does in a simulation.
France has recently been affected by several extreme weather events, with many households suffering property and vehicle damage due to high winds, flooding and hail. You may be owed compensation; here are more details. Harmful episodes are also becoming more common in France due to global warming. Depending on the type of incident, the process...
Tips to keep summer travel costs from skyrocketing An expert gave FOX 32 some tips on how to keep your vacation costs from skyrocketing. CHICAGO – The words ‘summer’ and ‘vacation’ go hand in hand. This year, the problem is that your hand may be digging deep into your wallet to pay for a trip....