🔬 optics

Selective avoidance of multiple line-of-sight obstacles at 130~m using locally bending space-time wave packets

This paper demonstrates that engineering the spatiotemporal spectrum of pulsed optical beams to form space-time wave packets enables them to locally bend around multiple line-of-sight obstacles over distances up to 130 meters, thereby selectively reaching designated targets that would otherwise be blocked.

Original authors: Layton A. Hall, Murat Yessenov, Isabelle Lebron, Ayman F. Abouraddy

Published 2026-07-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Layton A. Hall, Murat Yessenov, Isabelle Lebron, Ayman F. Abouraddy

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to throw a ball to a friend standing 130 meters away. But there's a problem: a giant, stubborn wall is blocking the direct path. If you throw a normal ball, it hits the wall and stops. If you try to throw a "magic" ball that curves around the wall like a boomerang, it might miss the wall, but it will also miss your friend because it keeps curving away instead of straightening out to hit the target.

Scientists at the University of Central Florida and other institutions have figured out a way to make a beam of light behave like a superhero that can do both: it can dodge a wall and still hit the friend on the other side. They call this a "locally bending space-time wave packet."

Here is how they did it, using a laser and a very smart mirror.

The Problem with Normal Curves

Usually, when scientists make light bend, they use something called an "Airy beam." Think of this like a car driving on a curved road. Once it turns, it stays on that curve. If you put a rock in the road, the car goes around it, but it never comes back to the straight path to reach your destination. The paper argues that this "fixed curve" approach isn't good enough if you want to hit a specific target behind an obstacle. You need a beam that can swerve locally—like a car that dodges a pothole and immediately straightens out to keep driving down the highway.

The Magic Trick: A Laser with a Twist

The team used a special laser that shoots out incredibly short pulses of light (only 100 femtoseconds long, which is a trillionth of a millionth of a second). They shot this laser through a diffraction grating (a surface with thousands of tiny lines) and then bounced it off a "Spatial Light Modulator" (SLM). You can think of the SLM as a high-tech, programmable mirror that can change the shape of the light's wave just by adding a little digital "phase" twist to it.

By carefully programming this mirror, they engineered the light so that its spatial shape (where it is) and its temporal shape (when it is) are locked together in a very specific way. This creates a "space-time wave packet" (STWP).

The Experiment: Dodging Walls in the Open Air

The researchers didn't just do this in a computer simulation; they took it outside to a massive laser-testing facility in Florida. They set up a path stretching 130 meters across an open field.

They tested two tricky scenarios using cardboard cutouts as obstacles:

The "Sandwich" Scenario: They placed a target in the middle, with one obstacle in front of it and another behind it. The goal was to hit the target while dodging both the front and back walls.
The "Double Trouble" Scenario: They placed two obstacles in a row before the target. The goal was to dodge both walls and still hit the target at the end of the line.

The Results: Light That Knows Where to Go

Using their specially programmed laser, they successfully created "nulls" (areas where the light intensity drops to almost zero) exactly where the obstacles were.

In the first scenario: The beam bent around the first obstacle, straightened out to hit the target, and then bent again to avoid the second obstacle.
In the second scenario: The beam bent around the first obstacle, then bent around the second, and finally hit the target.

When they measured the power hitting the targets, the results were clear. When they turned on the "dodge mode," the light hitting the cardboard obstacles dropped to about 25% of its original strength (meaning it mostly missed them), while the light hitting the target remained strong and steady. The beam essentially created a "shadow" or a gap right where the obstacle was, allowing the rest of the beam to flow around it and continue on its straight path.

What This Means (and What It Doesn't)

The paper suggests that this technique could be a game-changer for things like remote sensing, where you need to see past a tree or a building, or for "directed energy" applications where you need to hit a specific spot without burning the stuff in front of it. It could even help in medical treatments where a laser needs to reach a tumor without damaging the healthy tissue in the way.

However, the authors are careful to note that this is a new discovery. They suggest that while they proved it works over 130 meters in the open air, there is still a lot to learn. They don't know yet exactly how big an obstacle can be before this trick stops working, or how close two obstacles can be before the beam gets confused. They also mention that they used a "heuristic" (a smart guess-and-check method) to design the mirror's pattern, and they suspect that using machine learning in the future could make the beam even better at creating these gaps.

So, while we can't yet say this solves every obstacle problem in the world, the paper shows that for the first time, we have a laser beam that can play "dodgeball" with obstacles and still hit the bullseye, all while traveling in a straight line most of the time. It's a clever, locally bending trick that keeps the light on target.

Technical Summary: Selective Avoidance of Multiple Line-of-Sight Obstacles Using Locally Bending Space-Time Wave Packets

Problem Statement
Conventional optical beams propagate rectilinearly, making them susceptible to blockage by line-of-sight (LoS) obstacles. While self-accelerating beams (e.g., Airy beams) follow curved trajectories to avoid obstacles, they follow a fixed bending path. This fixed trajectory prevents the beam from reaching a designated LoS target located downstream of the obstacle, as the beam continues to curve away from the target axis. To address this, a beam is required that can "locally bend" around an obstacle—deviating from its straight path to bypass the obstruction—and then return to its rectilinear trajectory to strike the intended target. Furthermore, selectively avoiding multiple obstacles while maintaining incidence on a specific target represents a significant challenge that has not been experimentally verified with existing beam-bending technologies, particularly over large distances in open-air environments.

Methodology
The authors utilize Space-Time Wave Packets (STWPs), a class of spatiotemporally structured optical fields where spatial and temporal frequencies are tightly coupled. Unlike conventional pulsed beams with independent spatial and temporal bandwidths, STWPs are engineered by imposing a linear constraint (spectral tilt) between the temporal frequency and the axial wave number.

To achieve local bending, the researchers employ a phase-only modulation strategy using a Spatial Light Modulator (SLM). The core methodology involves:

Spectral Engineering: Modifying the spatiotemporal spectrum of a generic pulsed laser beam (center wavelength $\approx 1064$ nm, bandwidth $\approx 25$ nm, pulse width 100 fs) to create a specific relationship between spatial frequency and wavelength that varies along the propagation axis.
Algorithmic Design: An algorithm determines the SLM phase distribution $\Phi(x_s, \lambda)$ required to generate a beam trajectory $x_o(z)$ defined by a Gaussian-like local bend: $x_o(z) = \pm x_t \exp[-(z-z_1)^2/(2z_t^2)]$ . Here, $z_1$ is the obstacle location, $x_t$ is the transverse width of the avoidance region, and $z_t$ is the axial extent of the bend.
Experimental Setup: Experiments were conducted at the Townes Institute Science and Technology Experimentation Facility (TISTEF) in an open-air environment. The beam was magnified and propagated up to 130 meters. Obstacles were simulated using cardboard cutouts (25 mm wide) positioned on optical posts. A CCD camera measured the transverse intensity profiles at various axial locations.

Key Contributions

Demonstration of Local Bending: The paper demonstrates the first experimental realization of STWPs that locally bend around obstacles and return to a rectilinear path to hit a target, distinct from the fixed parabolic or power-law trajectories of traditional bending beams.
Multi-Obstacle Avoidance: The authors successfully engineered beams to selectively avoid multiple obstacles in a single propagation path. This includes scenarios where obstacles are placed both before and after the target, and scenarios with multiple obstacles preceding the target.
Open-Air Validation: The experiments were performed over distances extending to 130 meters in an uncontrolled, open-air environment, moving beyond the well-controlled laboratory scales typical of previous STWP studies.
Phase-Only Modulation: The obstacle avoidance is achieved through phase-only modulation of the spatiotemporal spectrum, avoiding the power losses associated with amplitude modulation techniques (such as shadow projection).

Experimental Results
The study validated two primary scenarios:

Scenario I (Target Between Obstacles): A target was placed at $z=50$ $z = 50$ m, flanked by an obstacle at $z=20$ $z = 20$ m (Obs1) and another at $z=80$ $z = 80$ m (Obs2).
- Step 1: A baseline propagation-invariant STWP confirmed alignment.
- Step 2: A local bend was introduced at $z=20$ m, creating an intensity null that reduced power at Obs1 by $\approx 75\%$ (to $\approx 25\%$ of initial) while maintaining full intensity at the target and Obs2.
- Step 3: A second local bend was added at $z=80$ m. The beam successfully avoided both Obs1 and Obs2, maintaining constant incidence on the target.
Scenario II (Obstacles Before Target): A target was placed at $z=130$ $z = 130$ m, with obstacles at $z=40$ $z = 40$ m (Obs1) and $z=100$ $z = 100$ m (Obs2).
- Similar to Scenario I, the beam was dynamically reconfigured to introduce local nulls at the obstacle locations.
- Measurements confirmed that the beam intensity at the obstacles dropped to $\approx 25\%$ of the baseline, while the target at 130 m received consistent, high-intensity incidence.
- The beam exhibited self-healing properties, recovering its profile after bypassing the obstacles to strike the final target.

Significance and Claims
The authors claim that these results demonstrate a viable method for selective incidence on targets in the presence of LoS obstacles. The significance lies in the ability to "sculpt" the beam path at will by tuning the spatiotemporal spectrum, allowing for the creation of on-axis nulls at prescribed locations without disrupting the beam's ability to reach a specific target.

The paper suggests that this technology could contribute to applications requiring selective targeting in:

Remote sensing and stand-off detection.
Directed energy systems.
Optical and radio-frequency communications where LoS connectivity must be maintained despite intervening obstacles.
Selectively delivered radiation therapies, where beams must navigate around sensitive tissue to reach a target.

The authors remain modest regarding the current state of the technology, noting that while the concept is verified, further work is needed to optimize the SLM phase design (potentially using machine learning), extend the operational range to 1 km, and determine the limits regarding obstacle size, axial separation, and the number of avoidable obstacles. They also note that future work could explore combining local bending with global bending or utilizing spectrally incoherent light sources.