Autonomous Driving Level Determination from Sensor and Road Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in determining the feasible function and level of autonomous driving due to the absence of detectable objects and varying sensor requirements, leading to inconsistent performance across different vehicles and environments.
Innovation Solution
An autonomous driving assistance device that includes a determination unit to assess the level of autonomous driving based on peripheral and road information, using tables to determine feasible functions and levels, and a control unit to perform autonomous driving accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the automatic driving assistance device determines that a pedestrian is likely to step onto the road based on recognition results from multiple sensors, then the risk of pedestrian accidents is reduced, but false detections may occur due to sensor errors or environmental factors
Solution Approach 1:
The system continuously monitors recognition results from multiple sensors and uses feedback loops to determine whether a pedestrian is likely to step onto the road. The control unit receives recognition results, determines pedestrian likelihood, and adjusts warnings or driving restrictions based on this determination, creating a closed-loop feedback system that improves detection reliability while managing complexity through systematic decision-making
Solution Approach 2:
The automatic driving assistance device integrates multiple sensor types (cameras, LIDAR, radar) that can serve multiple functions - detecting pedestrians, vehicles, road conditions, and environmental factors. This multi-functional approach allows the system to use the same sensor infrastructure for various detection purposes, improving pedestrian detection accuracy without proportionally increasing system complexity
2Reliability
If the control unit issues a warning to the driver or restricts driving when a pedestrian is likely to step onto the road, then pedestrian safety is improved, but driving efficiency is reduced due to frequent interruptions
Solution Approach 1:
The system applies partial action by issuing warnings or driving restrictions only when the determined likelihood of pedestrian presence exceeds a specific threshold, rather than responding to all detected pedestrians. This selective approach maintains pedestrian safety by acting on high-risk situations while avoiding unnecessary interruptions for low-risk cases, thereby preserving driving efficiency
Solution Approach 2:
The control unit dynamically adjusts the level of intervention (warning vs. driving restriction) based on the determined likelihood of pedestrian presence. When likelihood is below the threshold, no action is taken; when it exceeds the threshold, appropriate warnings or restrictions are issued. This dynamic response strategy optimizes the balance between safety and driving efficiency by matching the intensity of intervention to the level of risk
3Reliability
If the device continuously monitors recognition results to determine pedestrian likelihood, then detection reliability is improved, but energy consumption increases due to continuous sensor operation and data processing
Solution Approach 1:
The system implements periodic monitoring of recognition results rather than continuous monitoring at maximum intensity. The control unit determines pedestrian likelihood by periodically evaluating sensor data and adjusting its monitoring intensity based on environmental context and detected risk levels, thereby maintaining detection reliability while reducing overall energy consumption through intermittent rather than constant operation
Solution Approach 2:
The device changes operational parameters dynamically - adjusting sensor activation levels, data processing intensity, and monitoring frequency based on environmental conditions and detected risk. When pedestrian risk is low, the system operates in a lower-power mode with reduced monitoring intensity; when risk increases, it transitions to higher-power modes with more intensive monitoring, optimizing the balance between detection reliability and energy consumption
Data Source
Figure 1
Figure 2
Figure 3A~3C
AI summary
A driving assistance device 1 performs autonomous driving based on an output from a sensor unit 13 that acquires information on surroundings of a vehicle or information on the state of the vehicle. The driving assistance device 1 stores an autonomous driving determination table Tj based on autonomous driving compatibility information 24. By referring to the autonomous driving determination table Tj based on sensor information 23 on the sensor unit 13 and road element information Ie on a predetermined road section, the driving assistance device 1 determines a set of an autonomous driving function Fc and an automation level Lv that can be performed in the predetermined road section.