← All research storiesEngineering

Peer-reviewedEngineering

Robots coming up from behind seem faster than they really are, and that could change how they’re programmed

In five experiments, people judged robots approaching from behind as faster than identical robots coming from the front, a bias robot designers could plan around.

How far alongPeer-reviewedReviewed by independent experts for a journal before publication. Peer review is a quality check, not a grade of evidence strength: confidence still depends on study design, limitations, replication and the wider evidence.
How soon could it reach you?A few years outPromising, but needs more testing or engineering first. Our rough editorial guess is several years, with a lot of uncertainty.
Where it comes fromIEEE Robotics and Automation LettersToyohashi University of Technology · Sep 10, 2026Read the original →

What the scientists did

Delivery and cleaning robots are starting to roll through sidewalks, malls and hospitals. To feel safe, people need to judge how fast a robot is coming toward them.

Researchers ran five experiments using real self-driving robots and virtual reality. Robots approached people from the front or from behind at speeds between 0.5 and 1.0 meters per second (a slow-to-normal walking pace), and participants judged whether each robot seemed faster or slower than a reference speed. The team then added the bias they found to a computer model of how robots steer around people.

Words to know

Perception bias

A consistent way our senses misjudge something, even when we try to be accurate.

  • Thinking a robot is faster than it is because it’s behind you

Autonomous mobile robot

A robot that moves around on its own using sensors and software, without someone steering it.

  • Delivery robots
  • Warehouse robots
  • Robot vacuums

What they found

92% of participants consistently judged robots approaching from behind as faster than robots coming from the front at the same real speed. The effect wasn’t explained by body position, the robot’s looks, or how people moved; it seemed tied to what happens visually when you turn to look behind you.

In a computer simulation of a robot overtaking a person, using the largest bias the team observed, a bias-aware navigation model started avoiding the person up to 0.57 seconds earlier and kept up to 0.40 meters (about 16 inches) more distance. That is a simulated result, not a measurement from robots on real sidewalks.

92%of participants saw rear approaches as faster
0.40 mextra clearance, at most, in a simulation using the largest observed bias
0.57 searlier avoidance, at most, in the same simulation

Why it matters to you

If you’ve ever been startled by something coming up behind you, this study suggests your brain may make it seem faster than it is. Robots that account for that could feel less scary and safer to share a sidewalk or hallway with, especially for older people or kids.

Where this could lead

More comfortable delivery robots

Robots could slow down or give more space when passing people from behind.

Few years

Better design rules

Safety guidelines for public robots could cover how people perceive motion as well as physical distance.

Few years

These are possibilities the research points toward, not promises. Most early findings take years of testing before they reach everyday life, and some never do.

Why scientists care

Making robots physically safe isn’t enough; they also have to feel safe. Studies of human perception help engineers design machines people are comfortable living alongside.

Keep in mind

The tests used a limited range of speeds and specific robots, partly in virtual reality. The researchers say future work must check comfort, how people actually step aside, other speeds, and robots of different sizes and shapes. We read the university press release rather than the full paper; the paper title above is the one the release gives, and we will confirm it against the journal record.

How far along is the science?

  • Peer-reviewedReviewed by independent experts for a journal before publication. Peer review is a quality check, not a grade of evidence strength: confidence still depends on study design, limitations, replication and the wider evidence.
  • PreprintA manuscript shared publicly, often before or alongside journal review. Check its separate journal and review status (each story says what we know). Results may change.
  • Conference talkPresented to other scientists at a meeting. Usually short and early; a full paper may come later.
  • Agency reportPublished by a science agency such as NASA, NOAA or NIH. Reviewed internally, but not by an outside journal.
  • Press releaseAn announcement from a university or company. Useful, but always check the research it describes.

How soon could it reach you?

  • Could reach you soonCould show up in products, services or advice soon. Our rough editorial guess is about 1 to 3 years, not a forecast from the research.
  • A few years outPromising, but needs more testing or engineering first. Our rough editorial guess is several years, with a lot of uncertainty.
  • Long-term scienceBasic research that builds knowledge. Real-world uses, if any, are far off.
  • Small stepA small update to earlier work. Useful to researchers, but not a change you would notice yet.

Every story links to its original source: the paper itself, or the conference program when no paper is out yet. Explanations are written with AI, usually from the paper’s abstract (each story says what it was written from), then checked against the source. We only use journals, recognized preprint servers, science agencies, scientific conferences and established science news outlets.