Bathymetric survey — the systematic measurement of water depth to produce accurate seafloor and riverbed charts — has long been a labour-intensive, vessel-heavy operation. A traditional crewed survey campaign requires a dedicated research or work vessel, a full bridge watch, specialist hydrographers onboard, and the overhead of crew safety at sea for days or weeks at a time. The data quality delivered by that setup is high. The cost structure that comes with it, however, is equally substantial, and for a large share of the world's waterways that cost has historically priced out the survey entirely.

Uncrewed surface vessels (USVs) are changing that calculus. Small, remotely supervised autonomous platforms — ranging from catamaran hulls a metre or two long to larger diesel-electric craft capable of offshore operation — can run survey lines continuously without fatigue, require no onboard crew accommodation, and can be transported to remote sites by road or helicopter where a full survey ship could never reach. The economic implication is not simply that the same survey costs less. It is that surveys that were previously uneconomical become feasible, and surveys that were genuinely dangerous for a human crew can now be executed at arm's length.

The cost drivers in conventional hydrographic work cluster around three categories: vessel mobilisation and fuel, crew time (bridge watch, hydrographers, safety officers), and downtime caused by human factors such as fatigue limits, port calls, and weather windows sized for human comfort rather than sensor tolerance. USVs reduce or eliminate the first two and alter the third. A small electric or hybrid USV operating in sheltered or semi-sheltered water needs no port calls except for battery swaps. Its endurance on a single charge or fuel load may span a full working day without supervision. Running parallel survey lines — the methodical back-and-forth pattern hydrographers call "lawnmowing" — requires consistent speed and heading for clean swath overlap. Autopilot line-keeping on a USV is, in practice, more consistent than a helmsman managing watch fatigue at the end of a long shift.

The sensor payloads carried by modern survey USVs are, in most cases, the same instruments used on crewed vessels. A shallow-water multibeam echosounder mounted in the hull produces dense, co-registered depth soundings across a swath proportional to water depth. Single-beam echosounders offer a simpler, lower-cost alternative for corridors or projects where full-coverage multibeam is not required. Position and heading are provided by an inertial navigation system (INS) tightly coupled to RTK GNSS, giving centimetre-level horizontal accuracy when a correction network or base station is available. Sub-bottom profilers and side-scan sonar can be towed behind larger USV platforms when the project scope extends beyond depth-only measurement. The instrument list is not novel. What changes with USV deployment is the labour envelope around those instruments — fewer people required to operate and monitor them continuously.

Data pipeline and quality assurance deserve attention, because the autonomy of the survey platform does not remove the need for rigorous post-processing. Raw multibeam data is dense and noisy. Refraction corrections must be applied using sound velocity profiles measured through the water column, typically by a cast at the start and end of each survey day. Tidal corrections must reconcile observed water levels against chart datum. Motion compensation from the INS compensates for pitch, roll, and heave, but only to the accuracy limits of the motion reference unit. Each of these steps is the same whether the hull is crewed or not. What USV operations require, in addition, is that the remote operator monitoring the survey has enough situational awareness — live data feeds, system health telemetry, and a reliable communications link — to catch systematic errors before they contaminate an entire day's acquisition. The quality assurance workflow therefore shifts partly from the vessel to the shore station, and partly to automated flagging routines running on the edge processor onboard.

There are real limitations that a balanced discussion cannot omit. Sea state is the primary constraint. Most small USVs are rated for moderate conditions — typically significant wave heights well under two metres — because their motion characteristics in heavy weather degrade both sensor data quality and platform safety. Offshore surveys, exposed coastlines, and seasons with consistent swell remain challenging for the current generation of compact platforms. Larger USVs designed for offshore operation exist but approach the mass and cost of a small crewed vessel, narrowing the economic gap for blue-water work.

Regulatory environments add a second layer of complexity. Maritime autonomy regulations vary considerably by jurisdiction. Some national hydrographic offices and port authorities require a crewed safety vessel to accompany the USV in certain waterways. Collision regulations (COLREGs) impose obligations on any vessel underway, and how a USV demonstrates compliance — particularly the "proper lookout" requirement — is still being resolved in many jurisdictions. Operators working in busy shipping lanes, harbour approaches, or across national boundaries face a patchwork of permissions that a crewed survey vessel would navigate with a standard flag-state certificate.

Neither of these constraints is a permanent ceiling. USV hull designs and sea-keeping are improving. Regulatory frameworks for maritime autonomy are being actively developed by the IMO and by national maritime authorities in the UK, Norway, Singapore, and others. The trajectory points clearly toward broader operational envelopes over the next several years. The question for project planners today is not whether USVs belong in the hydrographic toolkit — they plainly do — but how to match the right platform to the specific water body, weather window, and data specification, and how to integrate the USV data stream into the same quality-controlled pipeline that certification bodies and engineering clients expect from any bathymetric deliverable.

The deeper shift is one that goes beyond any single survey campaign. As autonomous platforms accumulate operational hours and their reliability is demonstrated across a wider range of conditions, the engineering and regulatory communities will progressively extend the envelope of what they accept. That is the standard pattern for physical AI systems entering regulated industries: capability leads, acceptance follows, and the boundary between what requires human presence and what can run autonomously shifts in one direction only. Hydrographic survey is, in that sense, an early and unusually clear example of the broader transition underway.

水深测量长期以来是一项劳动密集、高度依赖船舶的作业。传统有人测量船需要专职值班人员、船上水文专家以及全程的海上安全保障。数据质量有保证，但成本结构同样可观，大量水道因此在历史上从未完成系统测量。

无人水面艇（USV）正在改变这一经济逻辑。小型自主平台无需船员驻守，可持续执行测线任务而不受疲劳限制，可被公路或直升机运抵调查船无法抵达的偏远区域，也可进入对人员存在切实危险的浅水或受污染水体。经济效益并非单纯的"同类任务成本更低"，而是令此前不可行的测量项目变得可行，令此前对人员高风险的测量可在安全距离外完成。

传统水文测量的成本集中在三类：船舶动员与燃料、人员时间（驾驶台值班、水文专家、安全人员），以及由疲劳限制、靠港补给与人员适应性天气窗口所造成的停工时间。USV 能够压缩或消除前两类，并重塑第三类。电动或混合动力 USV 在遮蔽水域作业时无需靠港，续航覆盖一个完整工作日。其自动导航在长时间值班末段的一致性，实际上优于人工操舵。

现代测量 USV 搭载的传感器——多波束测深仪、单波束测深仪、与 RTK GNSS 紧耦合的惯性导航系统——与有人船使用的仪器并无本质区别。真正改变的是围绕这些仪器的人力包络。数据后处理的严格性不因平台无人化而降低：声速剖面修正、潮汐改正与运动补偿均不可省略。质控中心从船上转移至岸基站与机载自动标记程序，但工作量并未消失。

当前的主要限制是海况。多数小型 USV 适用于有效波高不超过两米的环境，开阔海域或持续涌浪季节仍是挑战。监管层面同样存在复杂性：各司法管辖区对海事自主权的规定差异显著，COLREGs 对"正规了望"的要求在无人船语境下尚在各国逐步厘清。这些限制并非永久性的天花板——IMO 及挪威、新加坡、英国等国家海事机构正在积极推进自主航行框架的建立。

更深层的转变在于：随着自主平台积累运营时间，其可靠性在更宽泛的条件下得到验证，工程与监管界将逐步扩大可接受的作业范围。这正是物理 AI 系统进入受监管行业的标准路径：能力先行，接受度随后跟进，对人员在场的要求只会朝一个方向移动。水深测量，正是这一更大转型的早期典型案例。

水深測量長期以來是一項勞動密集、高度依賴船舶的作業。傳統有人測量船需要專職值班人員、船上水文專家以及全程的海上安全保障。數據品質有保證，但成本結構同樣可觀，大量水道因此在歷史上從未完成系統測量。

無人水面艇（USV）正在改變這一經濟邏輯。小型自主平台無需船員駐守，可持續執行測線任務而不受疲勞限制，可被公路或直升機運抵調查船無法抵達的偏遠區域，也可進入對人員存在切實危險的淺水或受污染水體。經濟效益並非單純的「同類任務成本更低」，而是令此前不可行的測量項目變得可行，令此前對人員高風險的測量可在安全距離外完成。

傳統水文測量的成本集中在三類：船舶動員與燃料、人員時間（駕駛台值班、水文專家、安全人員），以及由疲勞限制、靠港補給與人員適應性天氣窗口所造成的停工時間。USV 能夠壓縮或消除前兩類，並重塑第三類。電動或混合動力 USV 在遮蔽水域作業時無需靠港，續航覆蓋一個完整工作日。其自動導航在長時間值班末段的一致性，實際上優於人工操舵。

現代測量 USV 搭載的感測器——多波束測深儀、單波束測深儀、與 RTK GNSS 緊耦合的慣性導航系統——與有人船使用的儀器並無本質區別。真正改變的是圍繞這些儀器的人力包絡。數據後處理的嚴格性不因平台無人化而降低：聲速剖面修正、潮汐改正與運動補償均不可省略。質控中心從船上轉移至岸基站與機載自動標記程序，但工作量並未消失。

當前的主要限制是海況。多數小型 USV 適用於有效波高不超過兩米的環境，開闊海域或持續湧浪季節仍是挑戰。監管層面同樣存在複雜性：各司法管轄區對海事自主權的規定差異顯著，COLREGs 對「正規了望」的要求在無人船語境下尚在各國逐步釐清。這些限制並非永久性的天花板——IMO 及挪威、新加坡、英國等國家海事機構正在積極推進自主航行框架的建立。

更深層的轉變在於：隨著自主平台積累運營時間，其可靠性在更寬泛的條件下得到驗證，工程與監管界將逐步擴大可接受的作業範圍。這正是物理 AI 系統進入受監管行業的標準路徑：能力先行，接受度隨後跟進，對人員在場的要求只會朝一個方向移動。水深測量，正是這一更大轉型的早期典型案例。