Vehicle Camera Triangulation for Low-Speed Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to detect objects, especially people lying on the ground, and determine spatial mapping of a vehicle's surroundings when stationary or moving at very low speeds, which is necessary for safety requirements in driverless industrial vehicles.
Innovation Solution
A method using a single camera with an active actuator system to adjust the camera's position without altering the vehicle's driving state, combined with triangulation to determine object information, and optionally using multiple cameras for enhanced reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensors (ultrasonic, optical, radar) are used to detect objects in all directions, then detection reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent applies multi-functionality by using a single sensor system that can perform multiple detection functions. The sensor is configured to detect both the presence of objects and classify them into different object types (pedestrian, cyclist, animal, stationary object) using the same hardware component, thereby avoiding the need for separate specialized sensors for each detection task.
Solution Approach 2:
The patent utilizes parameter changes by varying the beam direction and detection parameters of a single sensor to achieve comprehensive coverage. The control unit adjusts the beam direction dynamically and changes detection parameters based on the detected object's characteristics, allowing one sensor to perform what would traditionally require multiple fixed sensors.
2Reliability
If sensor beams are distributed in all directions continuously, then detection coverage is improved, but energy consumption increases
Solution Approach 1:
The patent implements periodic action by scanning the sensor beam across different directions in a systematic sequence rather than continuously illuminating all directions. The beam is directed toward specific regions of interest based on periodic scanning patterns, reducing overall energy consumption while maintaining detection coverage.
Solution Approach 2:
The system performs preliminary action by using the first sensor to perform a preliminary detection sweep before activating the second sensor or before performing more detailed detection. This preliminary scanning allows the system to identify regions of interest and allocate energy more efficiently by focusing detailed detection only where needed.
3Reliability
If object detection is performed in all directions, then detection completeness is improved, but processing time increases
Solution Approach 1:
The patent applies segmentation by dividing the detection space into different regions and assigning different sensors to detect specific regions. The first sensor covers a first region while the second sensor covers a second region, allowing parallel detection that reduces overall processing time compared to a single sensor scanning all directions sequentially.
Solution Approach 2:
The system performs partial action by initially detecting only the most critical or prominent objects in the scene using reduced scanning coverage, then performing more complete detection only for regions containing detected objects. This approach achieves sufficient detection completeness for safety-critical applications while significantly reducing processing time.
Data Source
Figure 1~1a
Figure 2a~2b
Figure 3
AI summary
The invention relates to a method for determining object information relating to an object (O) in an environment (U) of a vehicle (1) with a camera (4), comprising at least the following steps: capturing the environment (U) with the camera (4) from a first position (SP1); changing the position (SP) of the camera (4); capturing the environment (U) with the camera (4) from a second position (SP2); and determining object information relating to an object (O) by selecting at least one first pixel in the first image and at least one second pixel in the second image, wherein the first pixel and the second pixel are selected in such a way that they are associated with the same object point on the object (O), and determining object coordinates of the associated object point by means of triangulation (T). According to the invention, the changing of the position (SP) of the camera (4) is brought about by actuating an active actuation system (8) in the vehicle (1), wherein the active actuation system (8) adjusts the at least one camera (4) along an adjustment path (W), without changing a driving status (Z) of the vehicle (1).