Continuous Safety Guarantees for Robotic Motion Planning

This direction is split into other directions: collision tunneling. Its last version stays readable and citable here.

Version 1 of 166 papers79 conceptsfrozen 2026-10-04

In traditional robotics, motion planning is often treated as a sequence of snapshots. A planner picks a series of waypoints, checks if the robot is colliding with an obstacle at each point, and—if the points are clear—declares the path "safe."

However, this discrete approach suffers from a fundamental flaw known as the tunneling effect. If a robot moves quickly or an obstacle is thin, the robot might be safe at time $t_1$ and safe at time $t_2$, but pass directly through a wall in the interval between them. To solve this, a growing body of research focuses on continuous-time safety, ensuring that the entire volume of space a robot sweeps through is free of obstructions.

The Geometry of Motion: Swept Volumes

The most intuitive way to guarantee safety is to stop thinking about where the robot is and start thinking about where the robot will be over an entire duration. This region of space is called the swept volume.

As Katherine A. Skinner, Ram Vasudevan, and Seth Isaacson demonstrate in Conformalized Reachable Sets for Obstacle Avoidance With Spheres, we can use neural networks to predict these volumes and provide rigorous probabilistic safety guarantees. By modeling the volume rather than discrete points, planners can navigate cluttered environments without the "false optimism" that a path is safe just because its waypoints are clear.

Recent advances have made these calculations more precise. J. Wang and colleagues in Implicit Swept Volume SDF: Enabling Continuous Collision-Free Trajectory Generation for Arbitrary Shapes introduced a method to calculate the exact Signed Distance Field (SDF) of a swept volume. This allows a planner to "push" a trajectory away from obstacles at every moment in time, effectively eliminating the risk of clipping thin objects.

From Waypoints to Continuous Functions

If the problem is discretization, the solution is to represent motion as a continuous mathematical function. Instead of a list of coordinates, researchers are using polynomials and series expansions to define trajectories that are inherently smooth and evaluable at any instant.

Certification and Formal Guarantees

For safety-critical systems, "likely safe" isn't enough. Researchers are applying formal methods to prove that a robot will never violate its boundaries.

One powerful approach involves Control Barrier Functions (CBFs). As Guang Yang, Calin Belta, and Roberto Tron argue in Continuous-time Signal Temporal Logic Planning with Control Barrier Function, discrete-time checks are fundamentally insufficient for autonomous systems. CBFs act as a mathematical "force field," minimally adjusting the robot's commands in real-time to ensure it stays within a safe set of states.

To make these proofs computationally feasible, Alexandre Amice and Russ Tedrake utilize Sums-of-Squares (SOS) optimization in Certifying Bimanual RRT Motion Plans in a Second. This allows them to mathematically certify that complex, high-degree-of-freedom trajectories—like a two-armed robot reaching into a shelf—are entirely collision-free in just a few seconds.

Balancing Safety and Speed

The challenge of continuous safety is computational cost. Checking every infinitesimal point in time is impossible. To bridge this gap, researchers use conservative advancement and adaptive discretization.

In Separation-Sensitive Collision Detection for Convex Objects, Jeff Erickson, Leonidas J. Guibas, and others proposed scaling computational effort based on distance. If objects are far apart, the system checks less frequently; as they approach, the "Kinetic Data Structure" updates more often.

Another strategy, seen in the CIAO framework, involves calculating an action radius. By knowing the robot's maximum possible velocity and acceleration, the planner can determine the maximum distance the robot could possibly travel between two discrete samples. If the robot maintains a safety buffer larger than this distance, it is mathematically impossible for a collision to occur in the "unseen" interval.

By shifting from point-based checks to volume-based and function-based guarantees, these researchers are moving robotics toward a future where "tunneling" through obstacles is a physical and mathematical impossibility.

Papers in this version

Showing 60 of 66 papers.

History & citationVersion 1 · 2026-10-04

Split into other directions · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_continuous_safety_guarantees_for_robotic_18a5828d,
  title = {Continuous Safety Guarantees for Robotic Motion Planning},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/continuous-safety-guarantees-for-robotic-motion-planning/rdr_18a5828dd8674a5a84233dcf59edda81},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_continuous_safety_guarantees_for_robotic_18a5828d_v1,
  title = {Continuous Safety Guarantees for Robotic Motion Planning},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/continuous-safety-guarantees-for-robotic-motion-planning/rdr_18a5828dd8674a5a84233dcf59edda81/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}