# Occupancy-Aware Neural Distance Perception for Manipulator Obstacle Avoidance in the Tokamak Vacuum Vessel

## TL;DR

This paper introduces a new neural framework for precise distance perception in robotic manipulation inside tokamak reactors, enabling real-time obstacle avoidance with high accuracy and speed.

## Contribution

The novel ONDP framework uses physically-stratified sampling and a lightweight neural network to enable high-frequency, sub-centimeter distance perception in constrained environments.

## Key findings

- The ONDP model achieves 2–3 mm accuracy near surfaces in tokamak vacuum vessel environments.
- The method supports high-frequency distance and normal queries at over 15 kHz for large-scale batches.
- ONDP offers a 5911× speed-up over traditional mesh-based methods, enabling real-time motion planning.

## Abstract

Accurate distance perception and collision reasoning are crucial for robotic manipulation in the confined interior of tokamak vacuum vessels. Traditional mesh- or voxel-based methods suffer from discretization artifacts, discontinuities, and heavy memory requirements, making them unsuitable for continuous geometric reasoning and optimization-based planning. This paper presents an Occupancy-Aware Neural Distance Perception (ONDP) framework that serves as a compact and differentiable geometric sensor for manipulator obstacle avoidance in reactor-like environments. To address the inadequacy of conventional sampling methods in such constrained environments, we introduce a Physically-Stratified Sampling strategy. This approach moves beyond heuristic adaptation to explicitly dictate data distribution based on specific engineering constraints. By injecting weighted quotas into critical safety buffers and enforcing symmetric boundary constraints, we ensure robust gradient learning in high-risk regions. A lightweight neural network is trained directly in physical units (millimeters) using a mean absolute error loss, ensuring strict adherence to engineering tolerances. The resulting model achieves approximately 2–3 mm near-surface accuracy and supports high-frequency distance and normal queries for real-time perception, monitoring, and motion planning. Experiments on a tokamak vessel model demonstrate that ONDP provides continuous, sub-centimeter geometric fidelity. Crucially, benchmark results confirm that the proposed method achieves a query frequency exceeding 15 kHz for large-scale batches, representing a 5911× speed-up over mesh-based queries. This breakthrough performance enables its seamless integration with trajectory optimization and model-predictive control frameworks for confined-space robotic manipulation.