Vehicle Dynamics Estimation Using Radar Lidar and Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle dynamics estimation systems, relying on MEMS IMUs and wheel speed sensors, face challenges such as increased errors during wheel slippage and unreliable GPS availability in urban environments, particularly in cluttered scenes, making them unsuitable for long-term dead-reckoning and accurate position and velocity determination.
Innovation Solution
An integrated system using low-cost MEMS IMUs in conjunction with radar, lidar, and camera sensors to correct vehicle dynamics through object tracking and Kalman filtering, which combines longitudinal and lateral state estimation processors to provide accurate vehicle speed corrections in real-time, even in conditions where GPS signals are unavailable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If MEMS IMUs and wheel speed sensors are used for vehicle dynamics estimation, then the system cost is reduced, but measurement precision deteriorates during wheel slippage conditions
Solution Approach 1:
The patent combines multiple sensing modalities (radar, lidar, camera, IMU, wheel speed sensors) into an integrated system that fuses data from all sources. This merging allows the system to maintain low cost while achieving high measurement precision through complementary sensing - radar provides accurate velocity during slippage, IMU provides continuous orientation, and wheel sensors provide normal operating data.
Solution Approach 2:
The patent introduces radar and lidar as intermediary sensors that measure stationary objects and use their known positions to calculate vehicle velocity independently of wheel rotation. This intermediary measurement path bypasses the wheel slippage problem, providing a reliable reference that mediates between the low-cost sensors and accurate measurement requirements.
2Measurement precision
If GPS is integrated with MEMS IMU to address bias and drift, then measurement precision is improved, but reliability deteriorates in urban canyons where GPS signals are unavailable
Solution Approach 1:
The patent applies different sensing strategies for different operational contexts. In open areas with GPS availability, the system uses GPS+IMU for high precision. In urban canyons where GPS is blocked, the system automatically transitions to radar/lidar-based dead reckoning. This local adaptation to environmental conditions maintains both precision and reliability across diverse scenarios.
Solution Approach 2:
The patent changes the operational parameters of the sensing system based on environmental conditions. When GPS signals are unavailable, the system switches from GPS-dependent modes to sensor fusion modes relying on radar, lidar, and IMU. This parameter change allows the system to maintain reliability by adapting to signal availability while preserving measurement precision through appropriate sensor selection.
3Device complexity
If a single camera is used for ego-motion estimation, then device complexity is reduced, but measurement precision deteriorates in cluttered scenes
Solution Approach 1:
The patent merges multiple sensing modalities (radar, lidar, camera) that each have different strengths. Radar provides accurate range and velocity measurements, lidar provides precise 3D spatial information, and camera provides rich visual context. This merging allows the system to maintain low device complexity while achieving high measurement precision through complementary data fusion that overcomes the limitations of any single sensor in cluttered scenes.
Data Source
AI summary
A system for estimation vehicle dynamics, including vehicle position and velocity, using a stationary object. The system includes an object sensor that provides object signals of the stationary object. The system also includes in-vehicle sensors that provide signals representative of vehicle motion. The system also includes an association processor that receives the object signals, and provides object tracking through multiple frames of data. The system also includes a longitudinal state estimation processor that receives the object signals and the sensor signals, and provides a correction of the vehicle speed in a forward direction. The system also includes a lateral state estimation processor that receives the object signals and the sensor signals, and provides a correction of the vehicle speed in the lateral direction.