Probabilistic Free-Space Mapping for Predicted Road Occupancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driving assistance systems lack information about predicted free/occupied space based on the behavior of other road users, leading to potential emergency interventions instead of gentle, early reactions, and fail to account for uncertainties in environment detection and road user prediction.
Innovation Solution
A probabilistic free space map is created using sensor data fusion and trajectory prediction of dynamic objects, incorporating predicted behavior and uncertainty, with a Bayesian network to optimize driving functions for comfort, safety, and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If current occupancy grid maps or object lists are used to represent free space, then the current state can be displayed, but information about predicted future occupancy based on road user behavior is missing
Solution Approach 1:
The patent applies preliminary action by predicting future trajectories of dynamic objects and pre-calculating their occupancy of grid cells ahead of time. Instead of waiting for objects to actually reach positions, the system projects their paths forward in time and marks grid cells as occupied based on predicted trajectories, enabling early detection of potential conflicts before they materialize in the current state map.
Solution Approach 2:
The patent implements dynamics by transitioning from a static occupancy grid map that only reflects current states to a dynamic probabilistic free space map that continuously updates predicted future occupancies. The system recalculates trajectories and occupancy probabilities at each time step, allowing the representation to adapt to changing road user behaviors and environmental conditions in real-time.
2Reliability
If emergency braking is applied in response to sudden merging detection, then safety is improved, but driver comfort deteriorates due to late reaction
Solution Approach 1:
The system performs preliminary action by detecting potential merging trajectories and calculating occupancy probabilities in advance before the merging vehicle actually enters the ego vehicle's path. When the probabilistic free space map shows high occupancy probability in future grid cells, the system can initiate gentle deceleration or early lane changes proactively, rather than waiting for sudden emergency conditions that require harsh braking.
Solution Approach 2:
The patent applies beforehand cushioning by creating a buffer of predicted occupancy zones that cushion against future collisions. The probabilistic free space map acts as a cushioning layer that anticipates potential conflicts and allows for smooth, gradual responses. By marking grid cells as occupied based on predicted trajectories, the system cushions the ego vehicle against sudden merges, enabling comfortable early reactions instead of late emergency interventions.
3Ease of operation
If low probability objects are ignored to reduce false alarms, then driver acceptance improves, but safety-critical emergency cases may be missed
Solution Approach 1:
The patent applies local quality by differentiating the treatment of grid cells based on their occupancy probability and spatial location. Instead of uniformly ignoring low probability objects, the system applies different quality thresholds to different regions: high probability cells trigger appropriate responses, while low probability cells are evaluated based on their context (e.g., whether they represent potential merging trajectories). This localized quality assessment allows the system to maintain driver acceptance by reducing false alarms in low-risk areas while preserving safety-critical detection in high-risk zones.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention relates to a method for creating a probabilistic free-space map with static (2a, 2b, 3) and dynamic objects (V1-V7) comprising the following steps: - Retrieving (S1) static objects (2a, 2b, 3) and a perception area polygon (WP) from an existing environment model; - Collecting (S2) predicted trajectories (T1, T2) of dynamic objects (V1-V7); - Combining (S3) the static objects (2a, 2b, 3), the perception area polygon (WP), and the predicted trajectories (T1, T2) in a first free-space map; - Defining (S4) a maximum prediction time; - Defining (S5) prediction time steps; - Setting (S6) a current prediction time and setting this current prediction time to the value 0 to set the start of a specified prediction period; - Setting (S7) confidence intervals (K) around the static (2a, 2b, 3) and dynamic objects (V1-V7);- Defining (S8) at least one uncertain area (U) around at least one static (2a, 2b, 3) or dynamic object (V1-V7); - Generating (S9) a first probabilistic free space map for the current prediction time; - Generating (S10) at least one further free space map for at least one prediction time step; - Evaluating (S11) the generated free space maps.;