Offboard Trajectory Replacement for Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in maintaining reliable trajectory calculation due to insufficient or poor-quality sensor data from environment sensors, leading to unsatisfactory user experience and potential safety issues when driver intervention is required.
Innovation Solution
A driver assistance system that calculates a trajectory using sensor data and can replace it with an offboard trajectory from a server when sensor data quality falls below a threshold, ensuring continuous autonomous control by requesting and receiving offboard trajectories proactively based on predicted conditions or critical areas identified by cumulative requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle relies solely on onboard environment sensors to calculate trajectories, then the system maintains simplicity and real-time responsiveness, but the reliability of autonomous driving deteriorates when sensor data quality is insufficient due to environmental factors such as missing lane markings, difficult visibility conditions, or snow-covered road surroundings
Solution Approach 1:
The patent introduces an offboard server as an intermediary to provide trajectory data when onboard sensors are insufficient. The server receives sensor data from the vehicle, processes it along with map data, and returns calculated trajectories. This mediator resolves the contradiction by externalizing the computational burden and data processing requirements, improving reliability without requiring the vehicle itself to become more complex.
Solution Approach 2:
The patent transitions from purely onboard sensor-based trajectory calculation to a hybrid approach that incorporates offboard computational resources. By adding the spatial dimension of cloud-based processing and combining it with onboard sensors, the system achieves higher reliability while distributing complexity across multiple locations and processing layers.
2Ease of operation
If the vehicle switches to manual driver control when sensor data quality is insufficient, then the immediate safety risk is reduced, but the user experience deteriorates due to loss of autonomous driving capability and requiring driver intervention
Solution Approach 1:
The system proactively requests offboard trajectories before onboard sensors become completely unreliable. By anticipating sensor failures and pre-fetching alternative trajectory data from the offboard server, the system maintains continuous autonomous operation without requiring driver intervention, thus preserving ease of operation while ensuring safety through multiple data sources.
3Reliability
If the system requests offboard trajectories proactively based on predicted conditions, then the availability of trajectories in difficult situations is improved, but the data transmission requirements and system complexity increase
Solution Approach 1:
The system optimizes data transmission by sending only the specific trajectory information that is locally needed and not already available. The offboard server provides targeted trajectory data for the vehicle's current location and predicted path, rather than transmitting complete map datasets or redundant information, thus improving trajectory availability while minimizing data transmission overhead.
Data Source
AI summary
A driver assistance system (10) for a vehicle includes a control unit (11), and a communication device (12) for receiving data from a server (30). The control unit (11) is configured to calculate a trajectory (T) for the vehicle on the basis of sensor data. The control unit (11) is also configured to replace the calculated trajectory (T) for the vehicle (20) with an offboard trajectory (T1) received from the server (30) under certain circumstances.