Grid-Based Environment Estimation Using Evidential Mapping and Particle Tracking

Abstract

Modeling and estimating the current local environment by processing sensor measurement data is essential for intelligent vehicles. Static obstacles, dynamic objects, and free space have to be appropriately represented, classified, and filtered. Occupancy grids, known for mapping static environments, provide a common low-level representation using occupancy probabilities with an implicit data association through the discrete grid structure. Extending this idea toward dynamic environments with moving objects requires a static/dynamic classification of measured occupancy and a tracking of the dynamic state of grid cells. In this work, we propose a new dynamic grid mapping approach. An evidential representation using the Dempster-Shafer framework is used to model hypotheses for static occupancy, dynamic occupancy, free space, and their combined hypotheses. These hypotheses are consistently estimated and accumulated in a dynamic grid map by an adapted evidential filtering, allowing one to distinguish static and dynamic occupancy. The evidential grid mapping is combined with a low-level particle filter tracking that is used to estimate cell velocity distributions and predict dynamic occupancy of the grid map. Static occupancy is directly modeled in the grid map without requiring particles, increasing efficiency and improving the static/dynamic classification due to the persistent map accumulation. Experimental results with real sensor data show the effectiveness of the proposed approach in challenging scenarios with occlusions and dense traffic.

Computationally Efficient Fail-safe Trajectory Planning for Self-driving Vehicles Using Convex Optimization

“…The set B ⊂ N + contains indices, referring to all safetyrelevant obstacles within the environment, e. g., obtained from on-board sensors of the vehicle [37]. In order to guarantee safety of planned motions, we assume the existence of a prediction which accounts for any feasible future motion of obstacles, see e. g., [38].…”

Efficient Computation of Invariably Safe States for Motion Planning of Self-Driving Vehicles

“…The lane-based environment W ⊂ R k of (1) is modeled as a subset of the Euclidean space. The set B ⊂ N contains indices referring to all safety-relevant dynamic and static obstacles, typically obtained using on-board sensors [24]. We use v ≥ 0 to denote velocities and a to denote accelerations.…”

Data Association for Grid-Based Object Tracking Using Particle Labeling

“…In [21], we proposed such a dynamic grid mapping approach that estimates evidence masses for the hypotheses static occupancy S, dynamic occupancy D, free space F , and their combined hypotheses. Hence, the unclassified occupancy hypothesis O = {S, D} is split into the individual hypotheses S and D, which are separately estimated.…”

“…1e. In addition to the dynamic grid mapping approach of [21], the predicted tracks are used to avoid that occupied cells of slow-moving objects converge from dynamic toward static in this work. Hence, dynamic evidence and thus the corresponding particles are retained in areas of the predicted tracks.…”

“…Recent approaches of dynamic occupancy grids use a gridbased particle filter to estimate the dynamic state of the occupied environment, e.g. [17]- [21]. Each particle represents a hypothesis of occupancy at a specific continuous position with a particular velocity.…”