Subject

Exoskeletons: wearable passive, hybrid and powered devices (rigid and soft) that assist or restore human movement, across medical rehabilitation, occupational, defence and consumer use

Landscape boundary

In: lower-limb, upper-limb and trunk devices; exosuits; control and AI methods; evidence, standards and reimbursement. Out: prosthetics, implanted neuro-interfaces and surgical robots, except where they cross over.

Date and version

14 November 2026  |  v1.0, desk-based first foresight

Foresight window

2027 to 2031 (baseline year: 2026). This is the Year n foresight: the 2027 foresight should take it into account.

Prepared for

Srijan Sanchar. Assumed decision-maker: a research funder or university research centre setting a five-year exoskeleton research agenda.

Method

IEF v3.0. Master equation: P(Sₙ₊₁ | CCF, M, P(Tᵢ|Sₙ), Ω, H) × V(Sₙ₊₁) = Strategic Forecast

Executive Summary

Exoskeleton research has moved from building machines that move people (1960s to 2000s) to engineering machines that infer and follow people. The landscape now splits in two. One track is commercial and device-led: self-balancing clinical systems, now consolidating (Wandercraft acquired Ekso Bionics on 2 October 2026), and powered consumer products (Hypershell's Halo, announced at IFA 2026 at about USD 2,299). The other track, mostly empty on the matrix, is evidence-led and human-led: co-adaptive therapy that moves from clinic to home, standardised long-term outcomes, and transparent assistance for ageing.

The deepest unresolved polarity across six decades is who leads the movement, the device or the human. The second is laboratory efficacy versus real-world adoption. Forecast state: Bifurcation. The tipping condition is a populated home-based adaptive therapy cell, backed by multi-site longitudinal trials, a standard outcome set and payer coverage, before proprietary installed-base data lock in the device-led path (roughly 2027 to 2030, a judgement).

Five interventions are proposed. The two earliest, a standard outcome and field-evidence framework and an open benchmark platform with consented data, can start within twelve months.

Review literature

AI methods for therapeutic and assistive exoskeletons and exosuits: intention recognition, reinforcement learning, digital twins, generative AI; need for patient-specific data and clinical validation (Science Robotics, 2025)

van der Kooij et al., TU Delft repository

Patent and publication signals

Rigid multi-joint motor-driven designs dominate; adaptive AI control is the fastest-growing cluster in 2022 to 2025; hybrid FES-exoskeleton rarely claimed in patents; thin RCT-level evidence; India has the highest patent count in that dataset

PatSnap Eureka (marketing page; dataset-limited)

Occupational evidence

49 studies (2018 to 2024); about 85% of devices passive; most trials two hours or less; poor fit for women and smaller bodies; social care has two studies

Southampton scoping review

Market

USD 541.6 m (2025) to USD 643.5 m (2026), 19% CAGR to 2034 per one vendor; medical 38.6% of revenue; lower body 56.8%; Asia Pacific fastest growing

Polaris Market Research

Industry events

Wandercraft acquires Ekso Bionics (2 Oct 2026); Ekso 2025 revenue down 29%; Medicare brace benefit code K1007; Aetna, UnitedHealthcare and Humana coverage reported; Hypershell Halo (2.6 kg, four powered joints, USD 2,299); Walk in New York clinic (10 Nov 2026)

HIT Consultant; Exoskeleton Report; Wareable; The Gadgeteer

Standards and regulation

ASTM F3795-25 risk-management guide (no safety requirements yet); FDA class II special controls; standards lacking; risk catalogue of falls, skin injury, fractures

ASTM; OPEdge review

India

AIIMS Delhi, DRDO and IIT Delhi gait-lab project for injured soldiers, early stage as of 2024

Digital Health News

Cognitive

The question has shifted from how to actuate a limb to how a device infers intent and co-adapts with a person. Methods now include machine-learned intention recognition, reinforcement learning with digital human twins, and human-in-the-loop optimisation of assistance.

Economic

A small market (about USD 0.6 bn in 2026 on one vendor's estimate) with high unit costs: personal medical devices have cost roughly USD 75,000 to 100,000 out of pocket, while consumer devices sell at USD 999 to 2,299. Value is captured through reimbursement (Medicare brace code K1007, private payers), employer injury-cost savings and consumer sales. Consolidation has begun.

Social

Exoskeletons are leaving the clinic for homes, workplaces and trails. Acceptance depends on comfort, fit (poor for women and smaller bodies), organisational support and voluntary use. Social care is almost absent from the evidence.

Epistemic

What counts as evidence is contested: lab metabolic-cost reductions (10 to 20% in soft systems; 18% in a seven-person post-stroke hip study) versus RCTs and long-term field studies, which are few. Most occupational trials run two hours or less, and outcome measures are not standardised.

Civilisational

Ageing populations and a projected 2.5 billion people needing assistive technology by 2050 (WHO figure cited by PatSnap) make mobility assistance a long-run capacity issue. At stake: independence across the lifespan versus dependence on proprietary, data-hungry wearables, and the normalisation of augmentation for able-bodied users.

1

Decision horizon

A research funder or centre must fix its five-year exoskeleton research agenda during 2027, for work running to 2031. (Assumed; replace with the actual decision.)

2

Strategic question

Where is exoskeleton research heading, which directions are under-explored, and which should be built so that the field serves people rather than only products?

3

Acceptable uncertainty

Act on a direction when its qualitative weight is Medium or higher and it passes the constraint test at least conditionally.

4

Provisional axes

A: Control authority, device-led → human-led. B: Deployment ecology, supervised clinic or lab → open everyday and work settings.

T1

1 Causal

Research direction is driven by three coupled forces: cheaper and lighter actuators, batteries and sensors; machine-learning control; and the economics of reimbursement and workplace injury. Evidence generation lags the commercial pull.

-

T2

1 Structural

The field is four loosely coupled communities (rehabilitation, occupational ergonomics, defence, consumer) with separate venues, metrics and regulators, sharing a core of biomechanics, control and actuation. About 85% of commercial occupational devices in the scoping review were passive.

-

T3

1 Comparative

The trajectory resembles hearing aids and e-bikes: assistive devices moving from prescription-only toward over-the-counter and self-selected use.

-

T4

1 Classificatory

The field partitions by actuation (passive, hybrid, powered), rigidity (rigid, soft), body region, and purpose (restore, assist, augment). Two dimensions could serve as axes: control authority and deployment ecology.

AXIS CANDIDATE x2

T5

2 Abductive

The gap between laboratory metabolic gains and slow adoption is best explained by comfort, fit, heat and organisational factors, not by weak assistive physics.

-

T6

2 Analogical

Learned control is converging on the speech-recognition pattern: models pretrained on large gait datasets, then personalised on the device.

-

T7

2 Combinatorial

Exoskeleton x functional electrical stimulation (hybrid FES-exoskeleton): a stated clinical priority with few patent claims, placed at shared control in supervised settings.

MATRIX SEED R3C1

T8

2 Combinatorial

Exosuit x smartphone gait sensing x tele-rehabilitation: a home-based adaptive therapy loop with clinician oversight.

MATRIX SEED R3C2

T9

2 Combinatorial

Exoskeleton x legged-robot locomotion algorithms (Wandercraft reuses its locomotion algorithms in an industrial humanoid): cross-fertilisation between wearables and robots.

MATRIX SEED R2C3

T10

2 Hypothetical

If metabolic cost is the wrong objective, the optimisation literature has tuned a proxy. Users may value stability, comfort and confidence more; research should optimise perceived effort and balance.

-

T11

3 Normative

Research should prioritise people for whom mobility is lost or declining (spinal cord injury, stroke, older adults, users in low-resource settings) before augmentation of the able-bodied.

-

T12

3 Skeptical

Forecasts differ widely between vendors, consumer launches serve a growth narrative, and 'AI' is a marketing label. Only a few clinical results are RCT-level.

-

T13

3 Deconstructive

Much exoskeleton research functions as a demonstration of control engineering (small samples, novelty rewards) more than as health-outcome research.

-

T14

3 Evaluative

The better future evaluates assistance on standardised, long-term, field-based outcomes that are comparable across devices and settings.

-

T15

4 Feedback

Reinforcing loop: installed-base data improves learned control, which improves outcomes, which grows the base. Balancing loops: unit cost and reimbursement slow the base. Consolidation concentrates the data.

-

T16

4 Emergent

At the FES, soft-robotics and AI interfaces, wearable neuro-robotic therapy emerges; at the consumer interface, the exoskeleton becomes a personal mobility product.

-

T17

4 Threshold

Thresholds: powered four-joint mass near 2.6 kg and price near USD 2,300 (crossed in consumer products, 2026); multi-payer reimbursement (K1007 and private payers, partly crossed); a published exoskeleton safety standard (not yet crossed: ASTM F3795-25 is a guide without requirements).

-

T18

4 Evolutionary

Selection favours light, soft or hybrid, intent-aware systems with low set-up burden. Rigid twelve-motor systems persist where balance is the need.

-

T19

5 Historical

Cycles: amplification ambition (1960s), rehabilitation robotics (1990s and 2000s), commercial clinical devices (2010s), then soft and passive occupational devices and AI-driven and consumer devices (2020s).

-

T20

5 Prospective

Baseline to 2031: clinical devices stay small and consolidated; consumer and occupational products expand; research attention concentrates on adaptive AI control, the fastest-growing cluster in 2022 to 2025.

-

T21

5 Constraint

Non-negotiables: user safety (falls, skin and tissue injury, fractures), human override, and privacy of body and movement data. These forbid device-prescribed motion outside supervised settings and opaque learned control with no safety envelope.

FORBIDDEN R1C2, R1C3, R1C4

T22

5 Scenario

Four structurally distinct futures: co-adaptive clinic-to-home care (R3 to R4, C1 to C2); co-adaptive work and everyday augmentation (C3 to C4); device-led clinical restoration (R2C1, R2C2, today's state); device-prescribed autonomy in the open (forbidden).

MATRIX QUADRANT x4

T23

5 Scenario

Fragmented landscape: incompatible metrics and no shared safety standard, so evidence cannot be compared and regulators default to caution.

MATRIX QUADRANT

Taxonomy Completeness Check (v3.0)

All eight checks pass.

NODE 1: Amplification versus control (1890s to 1970s)

Prior worldview

Machines could multiply human strength; the human was a pilot inside a powered frame.

Polarity

Machine power and autonomy versus human control and stability.

Failed attempts

Full-body powered frames of the 1960s (the General Electric Hardiman lineage) were heavy, power-limited and unstable. Early active orthoses for paraplegics (the Belgrade school) established walking control theory.

Synthesis

Bounded, human-in-the-loop assistance; stability and balance treated as first-class design problems.

CCF constraint

Any future synthesis must keep the human able to lead and override, and must be stable under real loads.

Axis Polarity Score

5: control authority recurs at every later node.

Political filter

Defence funding set early priorities toward strength and load carriage; therapeutic and disabled users were secondary.

NODE 2: Robot-led versus patient-led therapy (1990s to 2000s)

Prior worldview

Gait therapy means moving the limb precisely along a template.

Polarity

Rigid position control versus the patient's own effort and neuroplasticity.

Failed attempts

Fixed-trajectory driven orthoses; passive patient riding the machine.

Synthesis

Assist-as-needed and impedance control; rehabilitation robots (treadmill-based trainers) and early wearable assist devices.

CCF constraint

Any future synthesis must engage the user's own neuromuscular effort in therapy, and not replace it.

Axis Polarity Score

4

Political filter

Clinic-based, therapist-supervised models excluded home and low-resource users.

NODE 3: Mobility device versus clinical and economic proof (2010s)

Prior worldview

A working walking exoskeleton is its own proof of value.

Polarity

Restoration technology versus evidence of benefit and cost.

Failed attempts

High-cost personal devices with limited outcomes data and heavy set-up.

Synthesis

Regulatory pathway (FDA class II with special controls), clinical use, and later a reimbursement route (Medicare brace code K1007; private payers).

CCF constraint

Any future synthesis must show benefit beyond the lab on standard outcomes to earn regulatory and payer acceptance.

Axis Polarity Score

3: partly resolved by coverage, evidence still thin.

Political filter

Reimbursement rules define who counts as an eligible user.

NODE 4: Lab benefit versus real-world usability (mid-2010s to 2020s)

Prior worldview

A measured reduction in metabolic cost proves a device works.

Polarity

Laboratory efficacy versus comfort, fit and adoption.

Failed attempts

One-size frames; lab-only validation; short trials of two hours or less.

Synthesis

Soft exosuits (from the DARPA Warrior Web programme onward), human-in-the-loop optimisation of assistance, and a large passive occupational segment.

CCF constraint

Any future synthesis must be tolerable over a full day and fit diverse bodies, and must be validated in the field.

Axis Polarity Score

4

Political filter

Fit data skewed to male bodies; women and smaller users were underserved.

NODE 5: Learned autonomy versus verifiable safety (2020s)

Prior worldview

Better algorithms alone will make assistance transparent and safe.

Polarity

Machine-learned autonomy versus verifiable safety, user control and data governance.

Failed attempts

Not yet resolved. Risk guidance exists (ASTM F3795-25) but safety requirements are pending.

Synthesis (provisional)

Learned control inside an audited safety envelope, with consented use of movement data.

CCF constraint

Any future synthesis must bound learned behaviour inside a verifiable safety envelope, and govern body data.

Axis Polarity Score

5: it restates the Node 1 polarity at the level of AI.

Political filter

Data from installed bases accrues to a few firms, now consolidating.

C1

1

Any future synthesis of exoskeleton research must keep the human able to lead and override, and be stable under real loads.

Non-negotiable

C2

2

...must engage the user's own neuromuscular effort in therapeutic use, not replace it.

Provisional (therapy only)

C3

3

...must demonstrate benefit beyond the laboratory on standardised outcomes.

Non-negotiable

C4

4

...must be tolerable over a full day and fit diverse bodies.

Non-negotiable

C5

5

...must bound learned control inside a verifiable safety envelope and govern body data.

Non-negotiable

C6

3, 4

...must be affordable against reimbursement or employer return on investment.

Provisional

Full-body amplification suits for general use

Power density, mass and stability (Node 1)

Possible, through humanoid and 'physical AI' framing of heavy-duty suits

Fixed-trajectory gait driving

Undermines patient effort (Node 2)

Implicitly, through pretrained gait templates

One-size rigid frames

Fit and comfort (Node 4)

In some consumer and industrial products

Lab metabolic cost as sufficient proof

Weak link to adoption (Node 4)

Yes, in consumer marketing claims

Highest-scoring polarity

Nodes 1 and 5, score 5: who holds control authority over the movement.

Axis 1 (rows)

Control authority: device-prescribed → device-led with user triggers → shared and co-adaptive → human-led and transparent.

Check against Taxonomy flags

Agrees with the control-authority candidate in T4. The deployment-ecology candidate (Nodes 3 and 4, score 3 to 4) becomes Axis 2.

Deepest Polarity Rule

Remove the control-authority polarity and the history no longer makes sense: it is present at every node from the 1960s frames to learned consumer control.

1

Axis 1 (rows)

Control authority: device-prescribed → human-led (four levels, R1 to R4).

2

Axis 2 (columns)

Deployment ecology: supervised lab or clinic → clinic-to-home → structured workplace → open everyday and outdoor (C1 to C4).

3

Orthogonality test

High human authority in a supervised clinic is possible (assist-as-needed therapy). Device-prescribed motion in open settings is possible, if unsafe (a rigid walking frame outdoors). Both exist, so the axes are independent.

4

Granularity

4×4 for full analysis, because the decision horizon is five years and the landscape has four communities.

R1 Device-prescribed motion

Fixed-trajectory gait trainers

OCCUPIED (legacy), low V

T19

Fixed-trajectory rigid walker at home

FORBIDDEN: C1, C2

T21

Full-body amplification suit at work

FORBIDDEN: C1, C4, C5

Dead End Register

Prescribed motion in open terrain

FORBIDDEN: C1, C5

T21

R2 Device-led, user-triggered

Self-balancing clinical exoskeletons (Atalante X, EksoNR)

CURRENT STATE, Cluster Alpha, HIGH P

T2, T15, T20

Personal home exoskeletons for spinal cord injury (Eve, ReWalk Personal, Indego Personal)

CURRENT STATE, Cluster Alpha

T17

Powered task-triggered work assist

OCCUPIED (emerging)

T2, T9

Consumer powered hiking exoskeletons, preset modes

OCCUPIED, Cluster Beta

T16, T17

R3 Shared, co-adaptive

FES-hybrid co-adaptive neuro-rehabilitation exoskeleton

EMPTY > waystation (M1)

T7

Home tele-rehabilitation exosuit with gait sensing and clinician dashboard

EMPTY >> key waystation (M2)

T8

Personalised occupational exosuits (human-in-the-loop, intermittent use)

EMPTY (M3)

T5, T10, T14

Learned co-adaptive consumer exoskeletons (Halo generation)

OCCUPIED (frontier), Cluster Beta

T6, T16

R4 Human-led, transparent

Backdrivable open research platforms

EMPTY (M5)

T10

Transparent soft or hybrid assistance from clinic into home life, for rehabilitation and ageing

TARGET ★ (M4), V highest

T11, T14

Passive and hybrid occupational supports

CURRENT STATE (work), Cluster Gamma

T2, T5

Transparent augmentation for sport and outdoors

EMPTY (M6), values review

T11

CURRENT STATE

Gold

Yellow

Where a dominant programme sits now

TARGET ★

Green

Green

The cell the Directional Vector aims at

FORBIDDEN

Red

Pink

Violates a non-negotiable constraint (named in cell)

EMPTY / > >>

Standard

None / light blue

Unarticulated; light blue = waystation on the main trajectory

OCCUPIED

Standard

Grey

Populated by existing products or programmes, not on the main trajectory

1. Cluster pattern

Four clusters: Alpha (clinical device-led, R2C1 and R2C2, high weight); Beta (consumer powered and learned, R2C4 and R3C4); Gamma (passive occupational, R4C3); and a sparse, unformed Delta (co-adaptive clinic-to-home, R3C1, R3C2, R4C2). Alpha and Beta are device-led and product-driven; Delta is evidence-led. Pre-classification: Bifurcation.

2. Empty cell inventory

M1 FES-hybrid co-adaptive rehabilitation (R3C1). M2 home tele-rehabilitation exosuit (R3C2). M3 personalised occupational exosuits with intermittent use (R3C3). M4 transparent assistance for rehabilitation and ageing (R4C2, target). M5 open backdrivable research platforms (R4C1). M6 transparent augmentation for sport (R4C4). M1, M2, M4 and M5 are genuinely under-articulated; M6 needs a values review.

3. Forbidden zone geometry

Three cells in the top-right (R1C2, R1C3, R1C4): device-prescribed motion outside supervision. The feasible set is large. Watch for expansion: if safety requirements under a future ASTM or ISO standard and payer evidence rules tighten, R2C4 and R3C4 could become inadmissible for health claims. That is the Constraint-Collision signal to monitor.

4. Intervention trajectory

Main path: R2C1 → R3C1 → R3C2 → R4C2 (target). Occupational path: R4C3 → R3C3, adding co-adaptation to passive and hybrid supports. No forbidden cell is crossed. Sequence: shift Axis 1 first inside the clinic (R2 to R3 in C1) before moving Axis 2 toward home.

T5

Taxonomy

-

P

P

P

P

P

P

PASS

High

T7

Taxonomy

R3C1

P

P

C

C

C

C

CONDITIONAL

Medium

T9

Taxonomy

R2C3

C

P

C

C

C

C

CONDITIONAL

Low

T15

Taxonomy

-

P

P

P

P

C

P

CONDITIONAL

High

M2

III.0 matrix

R3C2

P

P

C

P

C

P

CONDITIONAL

Medium

M3

III.0 matrix

R3C3

P

P

C

C

C

P

CONDITIONAL

Medium

M4

III.0 matrix

R4C2

P

P

C

P

P

C

CONDITIONAL

Low

M5

III.0 matrix

R4C1

P

P

P

P

P

P

PASS

Low

M6

III.0 matrix

R4C4

P

P

P

C

C

C

CONDITIONAL

Low

Internal coherence

High

Medium (marketing claims outrun evidence)

High

High

Anomaly resolution

Low (cost and access unsolved)

Medium

Medium (fit and comfort unsolved)

High (evidence, access, adoption)

Constraint depth

Satisfies C1, C3 partly

Tension with C3, C5

Satisfies C1, C5; tension C4

Reinforces C1, C2, C3

Maturation

High (700+ centres)

Medium (units shipping from Nov 2026)

High

Low (seeds only; needs validation)

Overall weight

High

Medium to high

Medium to high

Medium

1

Fork

Path A, product-led: device-led clinical and consumer systems (R2C1, R2C2, R2C4, R3C4) scale with weak shared evidence. Path B, evidence-led: co-adaptive clinic-to-home care (R3C1, R3C2, R4C2) with standard outcomes.

2

Tipping condition

R3C2 becomes populated: a home-based adaptive therapy loop shown in a multi-site longitudinal trial on a standard outcome set, with payer coverage for it.

3

Intervention design

Standard outcome and field-evidence framework; FES-hybrid clinical programme; tele-rehabilitation pilots; open benchmark platform and consented dataset (Section V).

4

Timeline indicator

About 2027 to 2030 (judgement). The T15 loop saturates as consolidation (Wandercraft and Ekso, October 2026) pools installed-base data; once proprietary datasets set the performance benchmark, Path A becomes self-reinforcing.

H1 Abductive injection

Which empty cell, if populated, would change your understanding of exoskeletons most?

Physiatrists; rehabilitation engineers

H2 Values reassignment

Is R4C2 the right target, or should it be a different cell on ethical grounds (for example R4C4 augmentation)?

Disability advocates; ethicists; older users

H3 Early collision sensing

Is any forbidden cell actually feasible, for example supervised prescribed motion at home?

Regulatory affairs; safety engineers

H4 Suppressed thesis recovery

Should any Dead End Register idea return under current conditions?

Defence researchers; industry engineers

H5 Cross-domain isomorphism

Does this matrix resemble the hearing-aid or e-bike topology, or another domain?

Cross-field practitioners

1

Irreducible purpose

Enabling people to move and take part in life with independence and dignity across the lifespan.

2

Preservation conditions

Human agency and override; people's own muscular effort preserved in therapy; evidence-based, affordable access; fit for diverse bodies; verified safety; governed movement data.

3

Aim

This foresight is aimed at the future in which independent movement across the lifespan is preserved and expanded, under the condition that human agency, evidence, access, fit, safety and data governance are met. The target cell R4C2 embodies this aim.

Current state: R2C1, R2C2 (Cluster Alpha)

High

Medium

Probable baseline; access and cost limits

Prepare or redirect

Target: R4C2

Low to medium

Highest

The normative aim

Build toward

Waystations: R3C1, R3C2, R3C3

Lower

High

Path to target

Populate by intervention

Forbidden: R1C2, R1C3, R1C4

Eliminated

N/A

Outside the feasible set

Design against

INTERVENTION 1: Standard outcome and field-evidence framework

Matrix cell target

R3, C1 to C2 (enabling)

Trajectory position

Early

P×V gap addressed

Evidence gap (C3); comparability across devices (T23)

Mechanism

A consortium agrees a core outcome set (balance, falls, participation, comfort, perceived effort, adverse events, long-term use) and longitudinal field protocols, building on existing assessments such as the Exoworkathlon, and a common adverse-event register.

Resolution condition

Two or more trials and one regulator or payer cite the outcome set; observed in publications.

Timeline

Start 2027; first version 2028.

Actor

Funder, clinical and ergonomics societies, standards bodies (ASTM F48, ISO).

INTERVENTION 2: Open benchmark platform and consented shared dataset

Matrix cell target

R4, C1 (enabling); counters T15 data lock-in

Trajectory position

Early

P×V gap addressed

Data concentration after consolidation

Mechanism

A low-cost backdrivable reference exoskeleton (3D-printed actuator designs exist) and a consented movement dataset with governance for body data, available to academic and low-resource users.

Resolution condition

Independent groups publish results on the same platform and dataset.

Timeline

Prototype in 12 to 18 months.

Actor

Universities, technical institutes, open-hardware communities, funders.

INTERVENTION 3: FES-hybrid co-adaptive clinical programme

Matrix cell target

R3, C1 (M1)

Trajectory position

Early to mid

P×V gap addressed

Co-adaptation not yet clinically validated; few patent claims

Mechanism

Multi-centre trials of FES-exoskeleton hybrids with assist-as-needed learned control, using the Intervention 1 outcome set. India: build on the AIIMS Delhi, DRDO and IIT Delhi gait-lab collaboration (soldier mobility), extending to stroke and spinal cord injury populations.

Resolution condition

A multi-site trial reports on the core outcome set; hybrid claims appear in patents.

Timeline

Trial design 2027; recruitment 2028; results 2030.

Actor

Hospitals, rehabilitation institutes, defence and civil research agencies.

INTERVENTION 4: Home tele-rehabilitation exosuit pilot

Matrix cell target

R3, C2 (M2, the key waystation)

Trajectory position

Mid

P×V gap addressed

Tipping condition of Step III.4

Mechanism

Soft exosuit plus smartphone gait sensing plus clinician dashboard, with adaptive dose and a payer pilot, run inside the safety envelope required by C5.

Resolution condition

A payer or national scheme covers a home adaptive-therapy episode.

Timeline

Pilot 2028 to 2029.

Actor

Rehabilitation providers, payers, soft-robotics groups, device firms.

INTERVENTION 5: Participatory occupational co-design

Matrix cell target

R3, C3 (M3) from R4C3

Trajectory position

Mid

P×V gap addressed

Fit, comfort and underrepresented users (C4); social care evidence gap

Mechanism

Co-design with workers, including women and smaller-bodied workers and care staff; intermittent-use protocols; implementation frameworks for uptake; month-long field studies.

Resolution condition

Published long-term field studies in social care and manufacturing with adoption metrics.

Timeline

2027 to 2030.

Actor

Employers, insurers, occupational health researchers, unions.

1

Thesis-to-node consistency

Pass

Historical claims match Nodes 1 to 5; dates need primary-source check

2

Constraint-to-thesis traceability

Pass

C1 to C6 each trace to nodes; T21 and T17 condition theses

3

Forecast-to-taxonomy completeness

Partial

Delta rests mainly on Tier 2 and Tier 4 theses; add Tier 3 evidence

4

Vector-to-CCF coherence

Pass

Aim contradicts no constraint

5

P×V actionability

Pass

Interventions 1 and 2 executable within 12 months

6

Matrix-to-CCF consistency (III.0)

Pass, with a flag

Forbidden cells map to C1, C2, C4, C5. C6 forbids no cell: stated weakly, kept provisional

7

Trajectory feasibility

Pass

Main path crosses no forbidden cell

1

Phenomenon Definition and Objective Statement

Part I

Complete (decision assumed)

2

Thesis Inventory

Part II

Complete

3

CCN Map and CCF Document

Part III

Complete; dates to verify

4

Morphological Matrix

Part IV, III.0

Complete

5

Forward CCN Forecast Report

Part IV, III.1 to III.5

Human Nodes pending

6

Directional Vector and P×V Matrix

Part V, IV.1 to IV.2

Complete

7

Strategic Intervention Map

Part V, IV.3

Complete