Pittsburgh Dynamics provides AI-driven performance intelligence for electric vehicles, converting road‑surface perception, tire state, and chassis data into real‑time torque, braking, and stability decisions. Their onboard compute platform fuses vision, thermal cues, and sensor inputs to predict grip limits and optimize vehicle dynamics before the limit is reached, enabling higher usable performance at the tire.
Funding
Funding not disclosed
Founders
Product
Problem
Electric vehicles lack real-time intelligence that integrates road surface perception, tire condition, and chassis dynamics, leading to suboptimal torque, braking, and stability control, especially when grip conditions change rapidly.
Solution
Pittsburgh Dynamics provides an AI-driven performance intelligence platform that runs on onboard compute to fuse vision, thermal, tire‑state, and chassis data into a live grip‑aware model. The system predicts available traction ahead of the vehicle and adjusts torque distribution, braking force, and stability interventions before the limit is reached. By leveraging the precise command authority of multi‑motor EVs, the platform delivers adaptive driver assistance and autonomous‑ready chassis control without modifying the battery or chassis hardware. The solution continuously learns from sensor inputs to improve performance across varied road conditions, effectively extracting more usable performance from existing vehicle components.
Target Audience
Primary customers are performance‑oriented electric vehicle manufacturers and OEMs seeking to enhance chassis control and driver assistance systems without redesigning vehicle hardware.
Features
- Onboard AI inference engine that processes camera, thermal, and vehicle sensor data in real time
- Grip‑awareness model that predicts tire‑road interaction ahead of the vehicle
- Adaptive torque allocation across multiple electric motors based on predicted traction
- Integrated braking and stability control adjustments synchronized with torque decisions
- Continuous learning layer that refines predictions from ongoing driving data
- Compatibility with existing EV hardware, requiring no changes to battery or chassis architecture