Quantum computing,
domain by domain

Fusion couples three brutally hard physics regimes — plasma dynamics, materials science, and nuclear reactions — sub-problems where classical supercomputers are already nearing their ceiling. That motivates the interest in quantum approaches; it is not yet a demonstrated win.

To be blunt up front: no published result shows a quantum computer solving a fusion problem that classical machines cannot. This section separates genuine theoretical potential from the hard limits of the NISQ era.

Four research directions

• Plasma & MHD / Vlasov PDEsThe 6-D phase-space kinetic equations suffer dimensional blow-up. Quantum linear-systems (HHL) and Hamiltonian simulation offer theoretical exponential promise for such high-dimensional PDEs — but state-prep, readout, and conditioning costs currently eat most of the gain.
• First-wall materials & neutron damageDefect formation and transmutation in plasma-facing materials under 14 MeV neutron flux is a strongly-correlated many-body electronic-structure problem. VQE and quantum phase estimation map onto it naturally — but the logical qubits needed for chemical accuracy far exceed today's hardware.
• Nuclear reaction cross-sectionsNuclear-structure calculations for D-T and advanced fuel cycles are strongly-coupled few-body problems. Early quantum-simulation attempts target light nuclei and remain far from engineering-grade accuracy for fusion reaction rates.
• Coil-geometry optimizationStellarator coil and magnetic-configuration optimization is a high-dimensional non-convex combinatorial problem. QAOA and quantum annealing are candidates — yet classical optimizers and quantum-inspired tensor-network methods remain the practical workhorses today.

Sub-problem to quantum-method mapping

Classical HPC vs quantum on fusion sub-problems

Players & ongoing work

Timeline

• 1994Shor's algorithm The theoretical origin of the belief that quantum computers can deliver exponential speedups — and the backdrop for post-quantum cryptography and this page.
• 2000sTensor-network methods Classical quantum-inspired methods mature into practical compression tools for high-dimensional plasma and many-body problems.
• 2020sNISQ proof-of-concept era Small VQE and Hamiltonian-sim demonstrations on light nuclei and small molecules — still bounded by noise and scale.
• recentUKAEA quantum MOUs A fusion agency and quantum vendors establish formal exploratory partnerships — interest moves from papers to institutions.
• 2030sFault-tolerant horizon If thousands of logical qubits become real, electronic structure and some PDEs could enter the engineering-relevant regime — timing highly uncertain.

For fusion, quantum computing is a theoretically grounded mid/long-term research direction — not a near-term commercial hook. Honestly separating the two is itself the value we offer.

Largely exploratory today, concentrated on a handful of compute-intensive bottlenecks. In the in-car real-time perception and decision loop there is no deployable quantum advantage; the genuine commercial landing zone is post-quantum migration for V2X.

For autonomous driving, quantum computing is more a threat AVs must defend against than a tool that drives them. A vehicle a future CRQC can eventually break needs quantum-safe keys today.

Four exploratory directions

• Path planning & real-time combinatorial optimizationAnnealing and QAOA for fleet dispatch, routing and traffic-flow assignment — small-scale simulation only, far from in-car scale and latency.
• Sensor fusion & uncertaintyHigh-dimensional Bayesian inference and QML kernels; the Volkswagen + D-Wave traffic-flow and battery-materials work is back-office research, not in-car inference.
• NN training accelerationVariational quantum circuits (VQC) handle only toy subtasks today, with no material speedup for perception-model training.
• Cryptography & V2X securityThe reversed-logic direction: AVs need PQC to defend against quantum attack — the vertical market for the Quantum-Safe Scanner and Migrator.

Bottleneck → candidate quantum method → honest maturity

QC enabling AVs vs PQC defending AVs

Real players & drivers

Timeline

• 2017–2019Volkswagen + D-Wave run quantum traffic-flow optimization trials in Lisbon — proving back-office optimization, not in-car capability.
• 2020+C-V2X cellular vehicle-to-everything deployment scales up, widening the vehicle communications attack surface.
• 2022/2024UNECE R155/R156 cybersecurity and software-update regulations take effect for new vehicle types, making crypto compliance mandatory.
• 2024NIST finalizes ML-KEM/ML-DSA/SLH-DSA (FIPS 203/204/205), giving quantum-safe migration deployable standards.
• now →Long-lived fleets with 10–15 yr road life enter the migration window: keys shipping today must outlast a future CRQC.

Mostly motion planning, control-policy training, multi-robot coordination, perception, and security — but the one direction genuinely near commercial landing is not "robots using quantum computing," it is post-quantum migration of industrial-robot / AMR firmware and comms.

• 1. Motion planning & trajectory optimization. Inverse kinematics, obstacle avoidance and many-DoF path planning are at heart constrained-optimization problems. Annealing / QAOA have been tried on discretized path search — a theoretical edge on combinatorial sub-problems — but only at simplified simulation scale, nowhere near replacing mature classical planners (RRT*, MPC).
• 2. RL & control-policy training. QML / VQC could in theory accelerate parts of the linear algebra inside policy-gradient and value approximation, but qubit count and noise confine it to toy scale and offline simulation only — it does not enter the real control loop.
• 3. Multi-robot coordination & swarm. Warehouse / logistics AGV routing and task allocation are textbook combinatorial problems that map naturally to QUBO — the clearest theoretical edge, and the most realistic early landing in robotics. Yet solved instances stay small, and a measured win over classical heuristics is not yet established.
• 4. Sensor processing & SLAM. Localization and mapping lean on heavy matrix operations and probabilistic inference. Quantum linear algebra (HHL-class) has theoretical room for large linear systems, but HHL's preconditions are stringent and it is far from robotics' latency, power and form-factor budgets — not near-term feasible.
• 5. Cryptography & robot security (directly relevant). Industrial robots, cobots and AMRs routinely connect to factory networks and the cloud. Their firmware updates and device identity lean on RSA / ECC, now exposed to harvest-now-decrypt-later (HNDL) — adversaries capture encrypted traffic today and decrypt once quantum matures. Control systems need PQC to withstand a future quantum adversary. The logic is reversed: robots don't need quantum computing, they need post-quantum crypto.

Robots don't need to "use quantum computing" — their control systems need PQC to survive a future quantum adversary.

• 2018–Cobots & AMRs scale in smart manufacturing — connected-robot install base grows fast, expanding the security exposure
• 2018–Quantum-annealing AGV / scheduling research — QUBO-coded multi-robot routing and allocation become the liveliest quantum thread in robotics
• 2010s–IEC 62443 industrial cybersecurity adoption — OT-security standards spread across plants, mandating device auth and encrypted comms
• 2020China Cryptography Law & 密评 requirements — commercial-cryptography mandated; critical infrastructure must pass 密评, pulling robot makers into the customer pool
• 2024NIST PQC standards finalized — FIPS 203/204/205 — ML-KEM / ML-DSA / SLH-DSA finalized, giving firmware and identity migration a concrete basis
• 2025+Firmware migration windows for long-lived fleets — industrial robots serve 10+ years; under HNDL the migration window opens now, and earlier is safer

Quantum and communications actually split into three very different logic lines that get conflated constantly — worth untangling first: which one is just research, which is a separate hardware stack, and which actually connects to the business.

• 1. Network routing & resource allocation. 5G/6G spectrum allocation, base-station load balancing and network-slice scheduling are high-dimensional combinatorial problems. Operators (Vodafone, SK Telecom) have run small annealing/QAOA trials — limited scale, mostly proof-of-concept, with no evidence yet of a real edge over classical heuristic solvers.
• 2. Signal processing & channel estimation. Massive-MIMO channel estimation and beamforming involve large matrix operations; quantum linear algebra (HHL-style routines) has theoretical room, but state-preparation and readout overheads keep it off the deployment path — no commercial use exists yet.

• 1. Quantum key distribution (QKD). Uses no-cloning to distribute keys and can in principle detect eavesdropping. China has invested most (Micius satellite 墨子号, the Beijing–Shanghai backbone 京沪干线), but QKD has hard practical limits: dedicated fiber/satellite infrastructure, limited range (a few hundred km without trusted relays), high cost — and it only solves key distribution, not data encryption, signatures or authentication. It cannot replace PQC.
• 2. Quantum internet. Entanglement-based long-distance quantum-state networks remain at the lab / small-trial stage (quantum repeaters, error correction and other core problems unsolved) — far from practical use and irrelevant to near-term enterprise procurement.

• 1. Comms crypto is almost all RSA/ECC. TLS/HTTPS, VPN/IPsec, 5G/6G core-network authentication and satcom encryption — their handshakes and certificates run almost entirely on RSA/ECC. Once a CRQC (cryptographically relevant quantum computer) exists, those public-key links fall to Shor's algorithm — not weakened, but with private keys recoverable outright.
• 2. Harvest now, decrypt later (HNDL). Traffic intercepted today can be stored and decrypted in the future — a real threat to long-secrecy communications (diplomatic, military, financial, long-lived satcom links), and the clock is already running. This makes telecom operators, satellite communications and government/enterprise private networks the core PQC-migration customer base.

• 2016Micius quantum-science satellite launched — first satellite-to-ground QKD
• 2017Beijing–Shanghai QKD backbone completed (~2000 km, relying on trusted-relay nodes)
• 2023–24Cloudflare / Chrome roll out hybrid X25519+ML-KEM TLS at scale — PQC reaches real traffic
• 2024NIST finalizes FIPS 203 (ML-KEM) / 204 (ML-DSA) / 205 (SLH-DSA) — a firm standards baseline for comms migration
• 2024+GSMA post-quantum telco task force drives operator migration guidance; 5G network functions and 6G design windows become PQC entry points

Quantum computing and cryptography sit at both the threat and defense ends at once — this is the core thesis the entire post-quantum business rests on. The same physics that breaks RSA is what drives PQC standardization and migration.

• Shor — breaks public-key. Polynomial-time factoring & discrete log; once a large low-error CRQC exists, RSA / ECC / DH are fully broken and the private key is directly recoverable.
• Grover — weakens symmetric & hashes. Quadratic speedup: AES-128 → ~64-bit effective (still 2^64 ops, limited real threat), AES-256 → ~128-bit (safe). This is why NIST PQC focuses on public-key while symmetric just needs longer keys.
• Harvest now, decrypt later (HNDL). Intercept & store ciphertext today, decrypt later once a CRQC matures; this drives national timelines (US NSM-10, CNSA 2.0).

• Lattice. ML-KEM (ex-Kyber, KEM) and ML-DSA (ex-Dilithium, signatures), built on LWE; the mainstream path and the specialty of LAC co-designer Lu Xianhui 路献辉.
• Hash-based. SLH-DSA (ex-SPHINCS+); the most conservative assumption, relying only on hash functions, at the cost of large signatures.
• Code-based. HQC, selected by NIST as a backup KEM in March 2025; Classic McEliece dates back to 1978 and has survived decades of public analysis.
• Multivariate & isogeny — proven unreliable. Rainbow was broken over a weekend on a laptop in 2022; SIKE was broken the same year — the lesson: a quantum-safe label must survive long public cryptanalysis.

The duality thesis: the same physics that breaks RSA is why post-quantum migration is now a compliance deadline, not a research topic.

Algorithm families

Migration mapping

Players

Timeline

• 1994 Shor's algorithm — poly-time factoring & discrete log, the theoretical basis of the quantum threat.
• 2016 NIST PQC competition launched — open global public evaluation.
• 2022 Rainbow & SIKE broken — multivariate and isogeny paths eliminated.
• 2022 US NSM-10 & CNSA 2.0 — national migration timelines set.
• Aug 2024 FIPS 203/204/205 finalized — ML-KEM, ML-DSA, SLH-DSA standardized.
• Mar 2025 HQC backup KEM — non-lattice diversity for the lattice path.
• 2026 Implementation-layer vulnerabilities — Bernstein flags issues in several official ML-DSA implementations.