The framework, called PERSCO, was developed by Sanne van Waveren, Zulfiqar Zaidi, and Matthew Gombolay at the Georgia Institute of Technology. Its central insight is that perceived safety can be treated as a tunable control problem rather than a fixed design choice. The researchers build on control barrier functions, or CBFs, mathematical constructs that guarantee a robot stays within a set of safe states by constraining its control inputs at every time step. Traditional CBFs enforce physical safety with static parameters that never change during an interaction. PERSCO instead parameterizes the CBF with two variables that directly shape how the robot’s behavior feels to nearby humans: the minimum distance the robot must keep from each person, and the maximum deceleration it is allowed to use, which in turn caps how fast it may approach anyone.

These two parameters have intuitive physical meaning. The distance parameter defines an intimate space around each person that the robot may never enter. The deceleration parameter determines the stopping distance the robot must be able to achieve at its current speed; a robot permitted stronger braking can safely travel faster, because it can halt in a shorter distance. By adjusting the pair, the controller can make the robot behave anywhere from maximally cautious to maximally assertive. The key question is which combination a given person actually perceives as safe, and that is something no designer can hard-code in advance, because perceptions vary widely between individuals and contexts.

To answer it, PERSCO treats the problem as active learning. Each person carries a hidden perceived safety function that maps any parameter pair to a judgment of safe or unsafe. The robot cannot observe this function directly, so it maintains a surrogate model, a classifier trained on feedback, and probes the boundary between safe and unsafe parameter regions. Crucially, the researchers designed the feedback to be as unobtrusive as possible. Rather than asking people to fill out Likert scales mid-task, PERSCO adopts a principle of perceived safe until proven unsafe: humans only signal when they feel uncomfortable, using a simple visual cue, in this study a handheld AprilTag sign raised toward a camera. Silence is treated as implicit safe feedback, provided the robot has logged enough close encounters with that person without any complaint.

The learning algorithm is engineered to minimize how often people must intervene. When unsafe feedback arrives, the robot updates its classifier and selects the next candidate parameters using a novel sampling strategy that balances two criteria: entropy, which targets regions where the model is most uncertain about where the boundary lies, and diversity, which favors candidates far from previously tested ones. Importantly, the sampler only considers parameters the model predicts to be safe, so the robot never deliberately behaves in a way it believes will alarm the human. When no unsafe feedback arrives after repeated close encounters, the robot gradually relaxes its parameters, stepping toward the least restrictive pair on the safety boundary, which maximizes task efficiency while remaining at the edge of what the person tolerates.

All of this runs inside a model predictive control loop with a 0.1-second time step, where the perceived safety CBF is enforced over the entire planning horizon while accounting for predicted human motion. A separate, unchanging physical safety CBF with the most aggressive parameters guarantees collision avoidance at all times, so no matter how the learned parameters evolve, the robot can always brake to a standstill before reaching anyone. When parameter updates suddenly tighten the constraints and the robot temporarily finds itself outside the new safe set, a gradual recovery strategy using a slack variable steers it back smoothly; in simulation this reduced jerk by 25 percent and angular acceleration by a factor of 3.6 compared to abrupt corrections, avoiding the jarring sidesteps that quick recovery methods would produce.

Simulation experiments validated the technical choices. Among three candidate classifiers, a support vector classifier with a radial basis function kernel proved the clear winner, updating in about 1.3 milliseconds on average, fast enough for real-time control, while the neural network and Gaussian process alternatives exceeded the control loop’s time budget. Against a battery of classical active learning baselines and black-box optimizers, including multi-armed bandits and Bayesian optimization, PERSCO sampling achieved high accuracy in recovering ground-truth safety parameters while producing the lowest ratio of unsafe feedback events. In a simulated workplace with three moving pedestrians, the system converged to near-optimal parameters in roughly eight minutes, both with and without noise injected into the feedback.

The decisive test came with real humans. Fifty-four participants, organized into eighteen groups of three, performed a workplace-inspired assembly task, walking between workstations to place LED pins on breadboards while a Boston Dynamics Spot robot navigated the space autonomously, covering 7,074 meters over the course of the study. Participants experienced three conditions: individual adaptation, in which the robot learned separate parameters for each person; collective adaptation, in which one shared parameter set was updated from anyone’s feedback; and an adversarial condition, in which the robot responded to feedback by becoming more aggressive rather than more cautious. The adversarial condition served as a control to test whether adaptation itself, or only feedback-aligned adaptation, improves how safe people feel.

The results were striking. Both aligned conditions significantly outperformed the adversarial one on perceived safety, comfort, and anxiety, all with p-values below .001 and large effect sizes. Collective adaptation scored highest overall, and participants raised their feedback signs significantly less often under collective updates, suggesting that people benefit from feedback provided by their teammates. A mediation analysis revealed that the effect of condition on perceived safety was fully mediated by the average size of the robot’s safety boundary, meaning the psychological benefit flowed directly from the geometric changes in the robot’s enforced constraints. Notably, individual adaptation offered no task-performance advantage over collective adaptation, contrary to the researchers’ hypothesis, possibly because fewer parameter changes allowed the robot to plan more consistently.

The authors are candid about limitations: the study took place in a controlled environment, sessions were capped at twelve minutes so parameters did not always converge, and treating silence as safe feedback assumes people are attentive enough to complain when they feel threatened. Still, the work marks a meaningful shift in how roboticists think about safety. By framing perceived safety as a quantity that can be measured, learned, and optimized alongside task performance, PERSCO argues that true safety encompasses psychological well-being, not just the absence of collisions. As robots move into warehouses, hospitals, and factories, the systems that earn human trust may be the ones that ask, in effect, how their presence feels, and adjust accordingly.

Subject of Research: Perceived-safe control of mobile robots using active learning from human feedback

Article Title: PERSCO: Perceived safe control of mobile robots in human groups with active learning

Article References: van Waveren, S., Zaidi, Z., & Gombolay, M. (2026). PERSCO: Perceived safe control of mobile robots in human groups with active learning. Autonomous Robots, 50(4), Article 43. https://doi.org/10.1007/s10514-026-10262-7

Image Credits: AI Generated

DOI: 10.1007/s10514-026-10262-7

Keywords: perceived safety, human-robot interaction, control barrier functions, active learning, mobile robots, model predictive control, social robotics, human-aware motion planning, adaptive control, Boston Dynamics Spot, user study, machine learning

The core insight behind the work is that most existing fusion strategies are little more than a one-way handoff. A global planner produces a route, passes it to a local controller, and the two never speak again. The weights that govern how the local controller balances goal-seeking against obstacle avoidance are typically fixed at calibration time, which means a robot tuned for open corridors will oscillate at the mouth of a narrow doorway, and one tuned for tight spaces will waste energy making needless detours across empty floors. Cui’s framework closes that loop. A dynamic sub-goal tracking mechanism continuously feeds the geometric constraints of the global path into the local speed controller, while the local controller’s progress in turn influences how fast the robot advances along the global reference.

At the global layer, the traditional A* algorithm receives its most significant upgrade in the form of an obstacle potential field factor woven into the heuristic function. Classic A* relies on Euclidean or Manhattan distance to guide its search, which is efficient but geometrically blind: it happily hugs the edges of obstacles because those routes are nominally shortest. The improved heuristic multiplies the distance term by a factor of one plus a potential field penalty that grows as the inverse square of the distance to the nearest obstacle, but only within a preset safety threshold. The result is that the cost of a node rises sharply as it approaches an obstacle, pushing the planned path into open space. The author is candid about the theoretical trade-off: because the potential field term can inflate the heuristic value near obstacles, the modified A* no longer guarantees the globally shortest path, though completeness is preserved, meaning the algorithm will always find a collision-free route if one exists.

The raw output of any grid-based search is a jagged polyline of grid centers, riddled with collinear redundant nodes and tiny turns that would force a real robot into constant acceleration and deceleration. The framework addresses this in two stages. First, a bidirectional line-of-sight strategy, implemented with Bresenham collision checking, strips the path down to only its essential turning points. Second, those key points serve as control vertices for a cubic B-spline curve, which produces a trajectory with continuous second derivatives, known as C2 continuity. Unlike Bézier curves, B-splines have local support, so adjusting one control point only reshapes the curve locally. Because the spline approximates within the convex hull of its control polygon rather than passing exactly through the control points, the smoothed curve can occasionally bulge toward an obstacle. A posterior safety verification step samples the smoothed trajectory every 0.1 meters and, if any point comes within the robot’s 0.4-meter radius of an obstacle, contracts the local control points inward and regenerates the curve, falling back to the original polyline after five failed iterations. This rollback eliminated all collision risks in testing while retaining roughly 95 percent of the smoothing benefit.

The local layer is where the framework departs most clearly from convention. The dynamic window approach samples velocity pairs within the intersection of the robot’s hardware limits, its acceleration-constrained reachable speeds, and a safety region guaranteeing braking distance exceeds the distance to the nearest obstacle. Traditionally, candidate trajectories are scored by a fixed weighted sum of heading, clearance, and speed terms. Cui replaces those static coefficients with a two-input, two-output fuzzy inference system built on Gaussian membership functions. Obstacle distance and current velocity are fuzzified against a three-by-three rule base following a monotonic safety principle: the closer the obstacle and the higher the speed, the greater the avoidance weight. Centroid defuzzification then yields crisp weights each control cycle. In one documented scenario, as an obstacle closed from 5.0 meters to the 0.6-meter high-risk threshold, the avoidance weight surged from 0.2 to 0.95, allowing the robot to suppress its target-seeking instinct and pass through a narrow U-shaped passage where a fixed-weight controller oscillated helplessly.

The ablation experiments quantify each module’s contribution with unusual granularity. Adding the potential field factor increased path length by about 3.4 percent, the price of safety, but cut expanded search nodes by 30.3 percent. B-spline smoothing then reduced cumulative turning cost from 15.7 radians to 4.2 radians, a 73.2 percent reduction, while suppressing peak curvature from values as high as 10.0 per meter to below 0.45 per meter. The dynamic sub-goal mechanism, tested across 50 Monte Carlo trials, cut navigation time by 13.4 percent, reduced lateral tracking error by 65.4 percent, and lowered emergency stops by 75 percent compared with aiming directly at the final destination.

Head-to-head comparisons in a 100-by-100 grid scenario filled with irregular static obstacles and randomly moving dynamic disturbances pitted the framework against traditional A* plus fixed-weight DWA, artificial potential field fused with DWA, and RRT* fused with DWA, all tuned through grid searches over more than 200 parameter combinations each. The proposed method achieved a 96.0 percent obstacle avoidance success rate, 14 percentage points above the traditional baseline, and an average navigation time of 36.2 seconds, a 34.4 percent reduction. Path smoothness cost fell from 24.5 to 8.4. Statistical testing backed the margins: Cohen’s d reached 2.37 against the traditional method, and one-way ANOVA across navigation time, path length, and smoothness yielded F statistics between 18.7 and 35.2 with p below 0.001. Notably, the framework also outperformed reinforcement learning planners including TD3, SAC, and a hybrid SAC plus RRT configuration, beating the best of them by roughly 21 percent in navigation time without any of the 2-million-step training those methods require.

Real-time performance is a quiet triumph of the design. Local planning averages 11.6 milliseconds per step on an Intel i7-10750H, comfortably within the 20-millisecond budget for 50 Hz control, with memory demands under 10 megabytes, making deployment feasible on ARM-class embedded boards such as a Raspberry Pi 4 or Jetson Nano without GPU acceleration. Global replanning triggers only when obstacle displacement exceeds grid resolution, an average of 0.7 times per 100 control cycles, and costs about 162 milliseconds when it does. Every decision is traceable through an explicit fuzzy rule library, no pre-training or labeled data is needed, and the system can be dropped into a new environment directly, a combination of interpretability and deployability that black-box learners struggle to match.

Robustness testing under more realistic conditions used ROS Melodic and Gazebo 9.0 with simulated lidar noise, odometry drift, and tire slip. Success rate dipped from 96.0 to 90.0 percent and navigation time rose to 41.8 seconds, a decline the author characterizes as within acceptable engineering limits. A sensitivity analysis across four uncertainty sources showed positioning error most affected navigation time, lidar noise most affected success rate, and high measurement delay pushed success down to 80 percent, suggesting that filtering or prediction modules would be needed in high-noise settings.

The author is forthright about limitations. The framework is built for two-dimensional grid maps and differential-drive kinematics; extending it to Ackermann steering, omnidirectional platforms, aerial robots, or three-dimensional terrain would require recalibrated speed spaces and trajectory models. The fuzzy rules were handcrafted for a 0.4-meter-radius robot with a 1.5-meter-per-second top speed and may need retuning at other scales, and performance against fully adversarial moving obstacles remains untested. Future work targets physical prototype experiments, three-dimensional navigation with elevation data, and distributed multi-robot coordination. Even so, the study’s real contribution may be methodological: it demonstrates that the leap from loosely connected modules to deeply coupled integration, in which geometry, control, and perception feed one another in a closed loop, is what finally turns decades-old algorithms into a navigation system ready for the messy, unpredictable world outside the laboratory.

Subject of Research: Hierarchical mobile robot path planning integrating an improved A* algorithm with a fuzzy-adaptive dynamic window approach

Article Title: Robot movement path planning integrating A* algorithm and dynamic window

Article References: Cui, X. (2026). Robot movement path planning integrating A* algorithm and dynamic window. Discover Artificial Intelligence, 6(1), Article 1320. https://doi.org/10.1007/s44163-026-02379-6

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02379-6

Keywords: mobile robots, path planning, A* algorithm, dynamic window approach, fuzzy logic, B-spline, obstacle avoidance, autonomous navigation, trajectory smoothing, robotics, path smoothing, real-time control

The central insight behind the design is a marriage of two ideas that each solve half of the problem. The first half is memory. Conventional feedforward neural networks treat every moment as fresh, with no recollection of what happened a millisecond ago. For a robot rolling across uneven ground, that amnesia is costly, because the dynamics of the machine carry history within them: inertia, slipping wheels, lagging motors. The researchers address this by building their hidden neurons with self-feedback loops, a configuration known as a diagonal recurrent structure. Each hidden neuron feeds its own previous output back into itself, giving the network a lightweight form of memory without the sprawling cross-connections that make fully recurrent networks computationally expensive and notoriously difficult to train stably.

The second half of the problem is expressiveness. Standard neural controllers typically rely on fixed activation functions, such as sigmoids or hyperbolic tangents, to transform signals. The new controller instead employs polynomial activation functions, which can represent nonlinear relationships with remarkable flexibility while remaining analytically tractable. Polynomial functions have a long pedigree in control theory, and their smooth, well-behaved derivatives make them particularly attractive when the control algorithm must be proven stable rather than merely observed to work. By combining polynomial activations with diagonal recurrence, the hybrid architecture captures both the nonlinear character of the robot’s dynamics and the temporal dependencies that a pure feedforward design would miss.

What truly distinguishes the work, however, is its radical economy. Where conventional adaptive neural controllers may tune hundreds or thousands of parameters online, the proposed HP-DRNNC requires only six adjustable weights. This is not merely an aesthetic preference for minimalism. Every tunable parameter in an online learning system demands computation on every control cycle, and on the embedded processors that drive real robots, that computation budget is scarce. With six weights, the controller can adapt sample by sample, updating itself in real time as the robot moves, without the need for a pre-collected training dataset. The controller learns on the job, in the field, from the live behavior of the machine it is steering.

Efficiency alone is worthless if learning is unstable, and this is where the team leans on one of the oldest and most respected tools in control engineering: Lyapunov stability theory. Rather than choosing a learning rate by trial and error and hoping the network does not oscillate or diverge, the authors derive their adaptation law directly from a Lyapunov function, a mathematical construct whose decreasing value certifies that the system’s error energy is shrinking. By requiring that the Lyapunov function decrease along every learning step, they guarantee that the weight updates cannot destabilize the controller. On top of this guarantee, they derive an adaptive learning rate rule that optimizes the speed of convergence, allowing the network to learn as fast as stability permits and no faster. The result is a learning algorithm with a mathematical certificate of stability, not just empirical evidence.

To test the design, the researchers implemented the controller on a mobile robot tasked with executing multiple practical missions, the kind of mixed duty cycles that real deployments demand. The experiments subjected the robot to two of the most common enemies of control performance: mass uncertainty and external disturbance. Mass uncertainty arises naturally when a robot carries varying payloads, since the inertia the controller must overcome changes with every load. External disturbances, from friction variations to pushes and uneven terrain, inject errors that a rigid, pre-programmed controller cannot anticipate. The HP-DRNNC absorbed both challenges, re-tuning its six weights online and holding its tracking performance where lesser controllers drifted.

The numbers reported are striking. Across the practical missions tested, the proposed controller improved the performance indices over existing controllers by 51 percent in tasks involving mass uncertainty and by 40 percent in tasks involving external disturbance. Those are not marginal gains. In robotics, performance indices aggregate tracking error over a mission, so improvements of that magnitude translate into visibly tighter trajectories, shorter settling times and less wasted energy. The comparison against other established controllers suggests that the combination of recurrence, polynomial activation and Lyapunov-certified adaptation extracts more control quality per parameter than the heavier architectures it was measured against.

The implications extend well beyond one laboratory robot. Because the architecture is so light, it is a natural fit for low-cost microcontrollers and embedded systems, the computational environments where most commercial robots actually live. The same research group has a track record of realizing intelligent controllers on inexpensive hardware, and the six-weight design follows that philosophy to its logical conclusion. A controller that adapts sample by sample without a dataset also sidesteps one of the most tedious steps in modern robotics practice: collecting, curating and validating training data before deployment. For applications such as warehouse logistics, service robotics, agricultural automation and hospital delivery, where robots encounter conditions that no offline dataset fully anticipates, that online adaptability is a genuine operational advantage.

There is also a broader lesson in the work about the value of hybrid design in an era dominated by ever-larger neural networks. While much of machine learning races toward scale, this controller moves in the opposite direction, showing that careful architectural choices, grounded in classical stability theory, can shrink a network to a handful of parameters without sacrificing capability. The diagonal recurrent structure supplies memory, the polynomial activations supply expressive power, and the Lyapunov-derived learning law supplies mathematical assurance, each component covering a weakness of the others. For engineers designing the next generation of autonomous machines, the message is that intelligence in robotics is not only about bigger models. Sometimes it is about smarter ones, small enough to run on a chip, stable enough to trust with a moving vehicle, and adaptive enough to handle a world that refuses to sit still for a training set.

Subject of Research: A hybrid polynomial-diagonal recurrent neural network controller for stable adaptive control of mobile robots

Article Title: A stable and efficient hybrid polynomial-diagonal recurrent neural network controller for mobile robot applications

Article References: Hanna, Y. F., El-Nagar, A. M., El-Bardini, M., & Khater, A. A. (2026). A stable and efficient hybrid polynomial-diagonal recurrent neural network controller for mobile robot applications. International Journal of Intelligent Robotics and Applications. https://doi.org/10.1007/s41315-026-00591-2

Image Credits: AI Generated

DOI: 10.1007/s41315-026-00591-2

Keywords: mobile robots, recurrent neural network, polynomial activation, Lyapunov stability, adaptive control, trajectory tracking, nonholonomic robots, adaptive learning rate, embedded robotics, online learning, stable, efficient