Autonomous Vehicle Ladar Ground Plane Estimation

The patent implements a variable-size history buffer that dynamically adjusts its capacity based on vehicle speed. When the vehicle travels at high speed, the buffer expands to accommodate more range scans, ensuring sufficient data coverage for accurate ground plane estimation. When the vehicle travels at low speed, the buffer contracts to maintain appropriate data density. This dynamic adaptation resolves the contradiction by making the buffer size responsive to operational conditions rather than fixed, thereby maintaining measurement precision across varying vehicle speeds while preserving ease of operation through automated adjustment.

The patent employs a variable-size history buffer that automatically adjusts its capacity based on detected vehicle speed conditions. At high speeds, the buffer enlarges to capture sufficient spatial coverage area across multiple range scans, ensuring reliable ground plane estimation. At low speeds, the buffer reduces its size to maintain appropriate data density and avoid redundancy. This dynamic sizing mechanism enhances estimation reliability across diverse operating conditions while keeping the system relatively simple through automated speed-based adaptation rather than complex manual configuration.

The patent utilizes GPS/INS navigation data to transform range scan measurements into world coordinates, creating a universal reference frame that accommodates various vehicle states (stationary, moving, turning) and road configurations (flat, inclined, curved). This transformation enables the ground plane estimation algorithm to operate consistently across diverse conditions by expressing all measurements in a common coordinate system. The increased adaptability is achieved through this multi-functional coordinate transformation approach, which handles different vehicle dynamics and terrain types uniformly, justifying the additional processing requirements.