Adaptive Cruise Control Stationary Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing adaptive cruise control (ACC) systems fail to accurately identify and react to stationary objects, leading to uncomfortable acceleration towards parked cars or stopped vehicles, due to limitations in radar sensor technology.
Innovation Solution
A combined video and radar sensor approach is proposed, with a controller that uses vehicle parameter sensors, object detection sensors, and engine control systems to limit vehicle acceleration when a stationary object is detected within a predetermined threshold, preventing excessive acceleration and ensuring smooth vehicle operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radar sensors are used in ACC systems, then the system can detect objects and maintain cruise control functionality, but the system cannot accurately distinguish between vehicles and infrastructure such as poles, signs, or bridges
Solution Approach 1:
The patent introduces an intermediary classification mechanism that processes radar sensor data to distinguish between stationary objects (infrastructure) and moving objects (vehicles). This intermediary layer analyzes object characteristics and applies classification rules to determine whether detected objects are vehicles or infrastructure, thereby resolving the radar sensor's inability to accurately differentiate between them.
2Productivity
If the ACC system accelerates toward the stationary object to achieve the desired speed, then the system maintains its cruise control performance, but the driver experiences uncomfortable and potentially unsafe acceleration toward stationary objects
Solution Approach 1:
The patent applies preliminary anti-action by implementing a preventive measure that checks for stationary objects in the vehicle's path before allowing acceleration. When a stationary object is detected, the system preemptively limits acceleration to prevent the harmful effect of accelerating toward the object, thereby eliminating driver discomfort and safety risks before they can occur.
Solution Approach 2:
The system implements feedback by continuously monitoring the vehicle's path and detected objects, and adjusting acceleration commands based on this information. When stationary objects are detected within the predicted vehicle path, the feedback loop modifies the acceleration profile to limit acceleration, thereby preventing uncomfortable or unsafe situations while maintaining cruise control functionality.
3Measurement precision
If video sensors are used instead of radar sensors, then the system can better distinguish between vehicles and infrastructure, but the system exhibits weaknesses in estimating longitudinal velocities and accelerations
Solution Approach 1:
The patent applies parameter changes by switching the detection modality from radar to video sensors, which changes the physical parameters being measured. Video sensors provide superior optical resolution for object classification but require different processing approaches for velocity and acceleration estimation, utilizing frame-to-frame analysis and optical flow techniques to compensate for the sensor type's limitations.
4Measurement precision
If a combined video and radar sensor approach is used, then the system can accurately identify stationary objects and maintain good velocity estimation, but additional sensors are required which increases the cost of the system
Solution Approach 1:
The patent merges video and radar sensor systems into a unified detection framework. The video sensor provides high-resolution object classification and identification, while the radar sensor supplies accurate velocity and acceleration data. By combining the strengths of both sensor types, the system achieves superior stationary object identification and velocity estimation without relying on a single sensor modality.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An ACC system and object detection method for a vehicle. The ACC system includes a vehicle parameter sensor, an object detection sensor, and a controller. The controller is configured to calculate the vehicle's path based on a signal from the parameter sensor, detect an object based on the vehicle's path and a signal from the object detection sensor, determine an acceleration of the vehicle, and prevent the acceleration of the vehicle from increasing while the object is detected and the acceleration of the vehicle is greater than a predetermined acceleration threshold. The object detection method includes identifying stationary objects in a path of a vehicle and limiting an acceleration of the vehicle when the acceleration is above a threshold and a stationary object is identified.