Robotics · 2026-11-03

Scaling Trusted Autonomy: Towards Generally Capable Verifiable Robots

Benedict Quartey

arXiv:2611.00158PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

The authors argue that scaling trusted autonomy requires a neurosymbolic approach that unifies general learned capabilities with explicit symbolic structure for reasoning and verification. They propose three interfaces—LIMP for language-to-plan grounding and inspectable interpretation, REALM for evolving 3D scene state that is updated after predicted manipulation, and PACTS for composing visuomotor skills with predictions of symbolic effects.

Why it matters

The paper targets the capability–assurance gap in foundation-model robots: fluency or success rates do not guarantee correct interpretation, intended effects, or alignment between internal state and reality. The authors aim to make robot behavior assessable by exposing interpreted requirements, predicted action effects for monitoring, and an evolving world representation that can be reconciled with observations.

Method

  • LIMP grounds natural-language instructions in explicit task constraints and instruction-specific semantic maps, enabling task-and-motion planning while exposing the robot’s interpretation for inspection.
  • REALM maintains an editable photorealistic 3D Gaussian scene, predicts consequences of planned manipulation, and locally reconciles the scene so later instructions use an updated grounding and planning geometry.
  • PACTS learns visuomotor skills that jointly generate action trajectories and predictions of symbolic effects, supporting skill composition, progress monitoring, and recovery through replanning.
Abstract (from arXiv)

Robots powered by foundation models are increasingly able to interpret open-ended language, recognize unfamiliar objects, and operate across diverse environments. Yet broader capability does not establish that a robot has understood a request, that its actions will produce the intended effects, or that its internal representation still corresponds to reality. Classical symbolic planning provides explicit structure for composition and verification, but traditionally depends on hand-engineered skills and representations. Learned policies can broaden the robot's repertoire, but provide weak guarantees about what will happen and why. This thesis argues that scaling trusted autonomy requires a neurosymbolic approach that unifies the generality of learned models with symbolic structure for robust robot behavior. It explores three complementary directions. First, LIMP uses foundation models and spatial reasoning to ground natural-language instructions in explicit task constraints and instruction-specific semantic maps, enabling task-and-motion planning while exposing the robot's interpretation for inspection. Second, REALM maintains an editable, photorealistic 3D Gaussian scene, predicts the consequences of planned manipulation, and locally reconciles the scene so that subsequent instructions obtain their grounding and planning geometry from an updated state. Third, PACTS learns visuomotor skills that jointly generate action trajectories and predictions of their symbolic effects, supporting skill composition, progress monitoring, and recovery through replanning. Together, these contributions preserve explicit structure for reasoning and verification while broadening how it is obtained, providing a foundation for generally capable robots whose behavior can be assessed under the assumptions of their models.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First announced by arXiv: 2026-11-03 · Indexed by Pub-AI: 2026-11-03 UTC

preprint

Scaling Trusted Autonomy: Towards Generally Capable Verifiable Robots

Benedict Quartey

First announced by arXiv: 2026-11-03 · Original record ↗

arxiv
2611.00158
Explicit source versions (1)
  • v1 · Date not supplied
  1. Preprint available

    Source date: 2026-11-03 · First detected: 2026-11-03 UTC · Historical baseline

    Source evidence (1)
    • Official evidence ↗ · RSS guid=oai:arXiv.org:2611.00158v1; arxiv:announce_type=new; item pubDate (official metadata)
Abstract · arxiv

Robots powered by foundation models are increasingly able to interpret open-ended language, recognize unfamiliar objects, and operate across diverse environments. Yet broader capability does not establish that a robot has understood a request, that its actions will produce the intended effects, or that its internal representation still corresponds to reality. Classical symbolic planning provides explicit structure for composition and verification, but traditionally depends on hand-engineered skills and representations. Learned policies can broaden the robot's repertoire, but provide weak guarantees about what will happen and why. This thesis argues that scaling trusted autonomy requires a neurosymbolic approach that unifies the generality of learned models with symbolic structure for robust robot behavior. It explores three complementary directions. First, LIMP uses foundation models and spatial reasoning to ground natural-language instructions in explicit task constraints and instruction-specific semantic maps, enabling task-and-motion planning while exposing the robot's interpretation for inspection. Second, REALM maintains an editable, photorealistic 3D Gaussian scene, predicts the consequences of planned manipulation, and locally reconciles the scene so that subsequent instructions obtain their grounding and planning geometry from an updated state. Third, PACTS learns visuomotor skills that jointly generate action trajectories and predictions of their symbolic effects, supporting skill composition, progress monitoring, and recovery through replanning. Together, these contributions preserve explicit structure for reasoning and verification while broadening how it is obtained, providing a foundation for generally capable robots whose behavior can be assessed under the assumptions of their models.