Driver Assistance Collision Warning Using Multi-Complexity Object Tracking

The patent segments the ascertainment process into multiple independent methods (first ascertainment method and second ascertainment method) that operate in parallel. Each method processes sensor data independently to determine object position and movement vector, allowing the system to select results based on confidence levels without waiting for all computations to complete, thus reducing overall reaction time while maintaining detection accuracy.

The system performs partial computation by using only the first ascertainment method when confidence is sufficient, and only adds the second ascertainment method when needed. This partial action approach avoids the full computational burden of always running both methods, reducing average computational complexity and reaction time while maintaining reliability when necessary.

The patent divides the computational task into segmented ascertainment methods with different complexity levels. The first method uses simpler computations suitable for real-time operation, while the second method provides more accurate but computationally intensive results. This segmentation allows the system to manage computational complexity by selecting appropriate methods based on situational needs rather than always employing the most complex approach.

The system dynamically adjusts the level of computational complexity by selecting which ascertainment methods to execute based on confidence levels and situational context. This dynamic approach allows the computational complexity to vary adaptively rather than remaining statically high, optimizing the balance between detection reliability and processing demands in real-time operation.

The patent creates a virtual copy of expensive sensor functionality through software-based ascertainment methods that process data from cheaper ultrasonic sensors. Instead of physically installing expensive radar, lidar, or camera systems, the system uses computational algorithms to simulate the detection capabilities these expensive sensors would provide, achieving similar measurement precision with much lower hardware costs.

The system replaces expensive, complex sensor hardware with cheaper ultrasonic sensors combined with sophisticated signal processing. The ultrasonic sensors are inexpensive and can be easily manufactured, and while they have limitations, the multiple ascertainment methods compensate for these limitations to achieve adequate detection accuracy without the high cost of premium sensor technology.

The system performs partial computation by executing only the necessary ascertainment methods based on confidence levels. When the first ascertainment method provides sufficient confidence, the system stops there and does not execute the more energy-intensive second method. This partial action approach reduces average computational energy demand while maintaining measurement precision when needed.

The system dynamically adjusts computational energy demand by selecting which ascertainment methods to execute based on real-time confidence assessments. This dynamic adaptation allows the system to consume less energy during routine operations while allocating more computational resources when detection accuracy becomes critical, optimizing the balance between precision and energy consumption.