COMPREDICT ads

A Wheel Force Transducer is one of the most accurate — and most expensive — pieces of measurement hardware in vehicle development. 🛞 Mounted on each wheel of a prototype, it captures forces and torques in three directions, in real time. Indispensable for chassis design, durability testing, and motion control work. Also: heavy, fragile, painful to install, and around €500k for a fully instrumented vehicle. Our Virtual Sensor for WFT does the same job — 0% additional hardware. Pure software, running on signals already present in the vehicle, outputting forces and torques at each wheel. Validated under bumps, rough roads, high-speed manoeuvres, and real public-road driving. The implication is bigger than the cost saving: every series-production vehicle on the road can now generate the kind of data that, until recently, was only available from a handful of instrumented prototypes. Real customer usage stops being a black box. Big credit to the Vehicle Dynamics team for the depth of work behind this one. 🙌 Curious to see it in action — or interested in joining the team building it? 👉 https://lnkd.in/egYwYXFg #TheVirtualSensorCompany #WheelForceTransducer #VehicleDynamics #SDV #COMPREDICT

A question we get from engineers when we describe vTPMS: how accurate is it really, compared to hardware? We published the validation data. Real-word driving tests show mean absolute errors comparable to direct hardware TPMS sensors across all four wheels under real driving conditions. Puncture detection achieved within UNR 141-02 regulatory timeframes. Slow diffusion events — gradual pressure loss — detected reliably without false positives. If you're evaluating software-based TPMS alternatives or working on a similar integration challenge, the data is there to review. https://lnkd.in/eKpQHWjM #VirtualSensors #TPMS #VehicleEngineering #ChassisEngineering #AutomotiveAI #UNR141 #COMPREDICT #Whitepaper

We put COMPREDICT's WFT Virtual Sensor on a European sports car OEM's mid-engine V8 — 1,852 kg, RWD, driven across braking, steering pad, and full track laps. Then we asked a specific question: does adding a 6D IMU to the standard CAN input signals actually improve wheel force estimation? The answer was more specific than we expected. • Longitudinal forces (Fx): no meaningful change. Standard CAN already gives the model enough observability. • Lateral forces (Fy): no meaningful change. Same story. • Vertical forces (Fz): 29–37% RMSE reduction across all four wheels — concentrated in the 10–50 Hz band, where road-induced heave, pitch, and roll live. Why: vertical acceleration, roll rate, and pitch rate give the model direct observability of body-motion dynamics that planar signals cannot see. The practical read for OEMs updating their sensor architecture: standard CAN-based WFT virtualization holds up for longitudinal and lateral estimation. 𝐹𝑢𝑙𝑙 𝑐𝑎𝑠𝑒 𝑠𝑡𝑢𝑑𝑦 — 85 𝑚𝑖𝑛𝑢𝑡𝑒𝑠 𝑜𝑓 𝑡𝑟𝑎𝑐𝑘 𝑑𝑎𝑡𝑎, 24 𝑠𝑖𝑔𝑛𝑎𝑙𝑠 𝑟𝑒𝑐𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑒𝑑 𝑎𝑡 200 𝐻𝑧, 𝑃𝑆𝐷 𝑎𝑛𝑎𝑙𝑦𝑠𝑖𝑠 𝑜𝑓 𝑡ℎ𝑒 10–50 𝐻𝑧 𝑏𝑎𝑛𝑑: 𝒍𝒊𝒏𝒌 𝒊𝒏 𝒄𝒐𝒎𝒎𝒆𝒏𝒕𝒔. #VirtualSensors #VehicleDynamics #ChassisEngineering #WheelForceTransducer #SDV

"Before I can analyze anything, I spend 60% of my time just fixing data." — Sound familiar? We built 𝐃𝐚𝐭𝐚𝐏𝐢𝐥𝐨𝐭 because this shouldn't be the reality for automotive data scientists and calibration engineers. It's a SaaS platform that takes raw vehicle driving logs and turns them into clean, labeled, analysis-ready datasets — automatically. ✪ Automatic signal matching, quality diagnostics & one-click auto-fix ✪ Event detection & labeling: braking, ABS, kick-down, acceleration and more ✪ One-click export — ready for calibration workflows or ML model training ✪ TISAX certified · EU-hosted · Built for OEM-grade data security 𝐽𝑜𝑖𝑛 𝑒𝑎𝑟𝑙𝑦 𝑎𝑐𝑐𝑒𝑠𝑠 𝑎𝑛𝑑 𝑟𝑒𝑐𝑒𝑖𝑣𝑒 100 𝐺𝐵 / €1,000 𝑖𝑛 𝑓𝑟𝑒𝑒 𝑐𝑜𝑚𝑝𝑢𝑡𝑒 𝑐𝑟𝑒𝑑𝑖𝑡𝑠 — ℎ𝑒𝑙𝑝 𝑠ℎ𝑎𝑝𝑒 𝑡ℎ𝑒 𝑛𝑒𝑥𝑡 𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑎𝑢𝑡𝑜𝑚𝑜𝑡𝑖𝑣𝑒 𝑑𝑎𝑡𝑎 𝑡𝑜𝑜𝑙𝑖𝑛𝑔. — COMPREDICT DataPilot early access programme 👉 Apply for free early access: compredict.ai/saas #DataPilot #AutomotiveData #MLEngineering #VehicleDynamics #CalibrationEngineering #DataQuality #SaaS #SDV #AutomotiveAI

Most vehicles have no idea how much tire or brake life is left — until it's too late. We're changing that. COMPREDICT's 𝐓𝐢𝐫𝐞 & 𝐁𝐫𝐚𝐤𝐞 𝐖𝐞𝐚𝐫 𝐕𝐢𝐫𝐭𝐮𝐚𝐥 𝐒𝐞𝐧𝐬𝐨𝐫 monitors tread depth and brake pad condition in real time, using only existing CAN data already present in your vehicle. No hardware. No installation. • Real-time tread depth monitoring per tire with <0.5 mm accuracy • Continuous brake pad & disc assessment with end-of-life forecasting • In-vehicle alerts via infotainment — or direct to dealership backend • Driving style analysis for personalized tire & brake recommendations • Self-calibrating: adjusts automatically after tire changes or replacements 𝑅𝑒𝑛𝑎𝑢𝑙𝑡 𝐺𝑟𝑜𝑢𝑝 — 𝑎𝑐𝑟𝑜𝑠𝑠 𝑅𝑒𝑛𝑎𝑢𝑙𝑡, 𝐷𝑎𝑐𝑖𝑎 𝑎𝑛𝑑 𝐴𝑙𝑝𝑖𝑛𝑒 — 𝑟𝑒𝑙𝑖𝑒𝑠 𝑜𝑛 𝑡ℎ𝑖𝑠 𝑡𝑒𝑐ℎ𝑛𝑜𝑙𝑜𝑔𝑦 𝑡𝑜 𝑝𝑜𝑤𝑒𝑟 𝑖𝑡𝑠 𝑑𝑎𝑡𝑎-𝑑𝑟𝑖𝑣𝑒𝑛 𝑎𝑓𝑡𝑒𝑟𝑠𝑎𝑙𝑒𝑠 𝑠𝑡𝑟𝑎𝑡𝑒𝑔𝑦. 👉 Download our e-brochure or book a COMPREDICTOR demo: https://lnkd.in/epFpJD8m #VirtualSensors #PredictiveMaintenance #TireWear #BrakeWear #SDV #AutomotiveAI #OEM #Aftersales

𝐃𝐀𝐂𝐈𝐀 𝐡𝐚𝐬 𝐬𝐞𝐥𝐞𝐜𝐭𝐞𝐝 𝐂𝐎𝐌𝐏𝐑𝐄𝐃𝐈𝐂𝐓'𝐬 𝐕𝐢𝐫𝐭𝐮𝐚𝐥 𝐓𝐢𝐫𝐞 𝐏𝐫𝐞𝐬𝐬𝐮𝐫𝐞 𝐌𝐨𝐧𝐢𝐭𝐨𝐫𝐢𝐧𝐠 𝐒𝐲𝐬𝐭𝐞𝐦 𝐟𝐨𝐫 𝐬𝐞𝐫𝐢𝐞𝐬 𝐩𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐯𝐞𝐡𝐢𝐜𝐥𝐞𝐬 🤝 A major milestone for us — and a strong validation of what software-based sensing can deliver at scale. Conventional indirect TPMS tells drivers that something is wrong. COMPREDICT’s vTPMS goes further: it delivers absolute tire pressure values per wheel via the DACIA app, with no additional hardware required. For Dacia, this enables a more informative digital experience in affordable vehicles. For the industry, it demonstrates that modern software can already deliver tangible customer value on existing vehicle platforms. The solution is scheduled to enter production in selected DACIA vehicles this year. Full announcement in comments 👇 cc: Stéphane Foulard Dr. Rafael Fietzek Stefan Hassels Matthias von Blohn Stefan Werner Martin Zeller Guillaume Desnouvaux Ousama Esbel Barbara Zaparoli Cunha #VirtualSensors #TPMS #SeriesProduction #DACIA #AutomotiveAI #SDV #VehicleData #PredictiveMaintenance #COMPREDICT #SoftwareDefinedVehicle