Dynamic Object Location Prediction Using LSTM Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle control systems require high computing power to accurately predict the location of dynamic objects like pedestrians, which can deviate from linear movement, posing challenges in collision avoidance and prevention, especially for partially automated vehicles.
Innovation Solution
A control system that combines data from multiple sensors using a first program to create an object list with location, speed, and open route, and employs a second program for future projection using LSTM or social LSTM networks to improve prediction accuracy and response time, enabling the vehicle to anticipate potential collisions and adjust its systems accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high computing power is used to accurately predict the location of dynamic objects, then prediction accuracy is improved, but system complexity and computational resources increase
Solution Approach 1:
The system segments the prediction task by creating separate object lists for different time points (current time, future time points) and processing them through distinct program modules. This divides the complex prediction problem into manageable segments that can be processed independently, reducing overall system complexity while maintaining accuracy.
Solution Approach 2:
The system performs preliminary actions by generating object lists at current and multiple future time points before actual prediction is needed. By pre-processing and storing object data in structured lists with timestamps, the system reduces real-time computational burden while maintaining high prediction accuracy.
2Reliability
If high computing power is used to accurately predict the location of dynamic objects, then collision avoidance capability is improved, but processing time and energy consumption increase
Solution Approach 1:
The system uses periodic action by generating object lists at discrete time intervals (current time and predefined future time points) rather than continuous processing. This periodic structure allows the system to maintain high collision avoidance capability through regular updates while reducing overall processing time and energy consumption compared to continuous high-power computation.
3Measurement precision
If detailed object data (location, speed, open route) is stored for multiple time points, then prediction precision is improved, but data storage requirements increase
Solution Approach 1:
The system extracts only the essential data elements needed for prediction (location, speed, open route) and stores them in a structured object list format. By extracting and storing only these critical parameters rather than complete sensor data, the system achieves high prediction precision while minimizing data storage requirements.
Data Source
AI summary
A control system and a method for predicting a location of dynamic objects, for example, of pedestrians, which are able to be detected by the sensors of a vehicle. The control system includes a multitude of sensors and a processing system, which is configured to combine with a first program the objects that are detected by the multitude of sensors to form an object list, each entry of the object list encompassing the location, a speed and an open route for each of the objects, and the object list including a time stamp; and to determine with a second program for at least a portion of the dynamic objects an additional object list from a predefined number of object lists, the additional object list including a time stamp for a future point in time and encompassing at least the location of the dynamic objects.