Autonomous Vehicle Trajectory Validation With Heterogeneous Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing redundant systems in autonomous vehicles are not infallible, as faulty sensors or sensing schemes can result in unsafe planned motions, leading to increased computing power requirements and performance issues.
Innovation Solution
An autonomous vehicle system that utilizes two processing systems: a first system forms a main perception from all sensors and computes trajectories, while a second system forms an auxiliary perception from a subset of sensors to validate these trajectories, thereby reducing computational efforts and enhancing inconsistency detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full redundancy is implemented using all perception sensors in both processing systems, then system reliability is improved, but computational load and processing time increase significantly
Solution Approach 1:
The patent divides the perception sensors into two distinct groups: a first set of perception sensors used by the first processing system to generate primary motion commands, and a second set of perception sensors used by the second processing system to generate secondary motion commands. This segmentation allows each processing system to operate with a dedicated sensor subset, reducing overall computational load while maintaining redundancy for safety validation.
Solution Approach 2:
The second processing system performs partial perception using only a subset of sensors specifically selected for validation purposes, rather than processing all sensor data. This partial action is sufficient to detect inconsistencies in the primary motion commands without requiring full computational redundancy, thereby improving efficiency while maintaining safety.
2Reliability
If redundant motion planners process all sensor signals, then trajectory validation is enhanced, but the number of operations and computing power requirements multiply
Solution Approach 1:
The patent extracts specific sensor signals from the complete sensor dataset and directs them exclusively to the second processing system for validation purposes. The second set of perception sensors is selected to provide complementary information that enables effective trajectory validation without requiring duplication of the full sensor processing pipeline, thereby reducing computational complexity.
Solution Approach 2:
Different processing systems are assigned different quality requirements and processing depths based on their functions. The first processing system performs comprehensive perception and motion planning, while the second processing system performs focused validation using a subset of sensors. This local differentiation of processing quality optimizes overall system efficiency while maintaining adequate validation capability.
3Ease of operation
If homogeneous sensor sets are used in both processing systems, then processing consistency is maintained, but inconsistency detection capability is reduced
Solution Approach 1:
The patent deliberately creates asymmetry in the sensor configuration between the two processing systems. The first processing system uses a first set of perception sensors while the second processing system uses a second set of perception sensors, which may include different sensor types or configurations. This asymmetric arrangement enhances the ability to detect inconsistencies in motion commands by providing diverse validation perspectives, while still maintaining sufficient processing consistency through standardized validation algorithms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention is notably directed to an autonomous vehicle, e.g., an autonomous or semi-autonomous vehicle such as a self-driving car. The autonomous vehicle comprises a drive-by-wire (DbW) system (300), a set of perception sensors 21 - 24, such as lidars and cameras, and two processing systems, i.e., a first processing system (100) and a second processing system (200). The first processing system is configured to form a main perception based on signals from each of the perception sensors of the set, estimate states of the vehicle based on feedback signals from the DbW system, and compute trajectories for the autonomous vehicle based on the perception formed and the estimated states. The second processing system is configured to form an auxiliary perception based on signals from only a subset of the perception sensors, validate the computed trajectories based on the auxiliary perception formed, and cause to forward the validated trajectories to the DbW system. In other words, distinct perceptions are formed from overlapping sets of sensors, whereby one of the perceptions formed is used to validate trajectories obtained from the other. This requires less computational efforts, inasmuch as less signals (and therefore less information) are required to form the auxiliary perception. However, doing so is more likely to allow inconsistencies to be detected, thanks to the heterogeneity of sensor signals considered in input to the main and auxiliary perceptions. The invention is further directed to related methods and computer program products.