Systematic Review: Contact Force Regulation in Human–Robot Co-Manipulation

What it is

Preprint: Control Strategies for Contact Force Regulation on Unknown Surfaces in Assistive Human–Robot Co-Manipulation: A Systematic Literature Review — Peter Ralph Desir, Arno Stienen (Delft University of Technology). Posted on engrXiv, 1 August 2026, CC BY 4.0.

From the abstract: “Musculoskeletal disorders from repetitive surface-finishing tasks remain a leading cause of occupational disability. Robotic assistive devices can reduce operator muscular load, yet existing control strategies typically assume known surface geometry, decouple force regulation from human guidance, or treat all contact directions identically.”

The review screened the contact-force-control literature against five components (surface estimation, force control, human-guidance coupling, anisotropy, online operation) to map which combinations exist and where the gap is. The full audit trail — stage protocols, raw query exports, per-record screening assessments for all candidates, PRISMA number reconciliation, and the 71-paper score matrix with per-cell provenance — is public in the companion repository, including a runnable script (verify_matrix.py) that recomputes the manuscript’s coverage tallies from the published matrix.

Results (all numbers live here, verbatim)

• 71 studies included. Identification-to-inclusion funnel: ~12,600 → ~9,534 → 350 → 72 → 71.
• Dated search update funnel (12 July 2026 run): 140 → 125 → 113 → 14 → 10 eligible.
• Component coverage across the 71 papers: C1 25/35%, C2 49/69%, C3 44/62%, C4 38/54%, C5 30/42%.
• Key gap: “only 7 papers combine C1+C2+C4 at IMPLICIT level or above, and 0 of 71 combine all five components with genuine online surface estimation.”
• Searches: Scopus + IEEE Xplore, 8 January 2026; update run 12 July 2026; preprint posted 1 August 2026.