In this paper a centralized method for collision avoidance among multiple agents is presented. It builds on the velocity obstacle (VO) concept and its extensions to arbitrary kino-dynamics and is applicable to heterogeneous groups of agents (with respect to size, kino-dynamics and aggressiveness) moving in 2D and 3D spaces. In addition, both static and dynamic obstacles can be considered in the framework. The method maximizes a joint utility function and is formulated as a mixed-integer quadratic program, where online computation can be achieved as a trade-off with solution optimality. In experiments with groups of two to 50 agents the benefits of the joint utility optimization are shown. By construction, it's suboptimal variant is at least as good as comparable decentralized methods, while retaining online capability for small groups of agents. In its optimal variant, the proposed algorithm can provide a benchmark for distributed collision avoidance methods, in particular for those based on the VO concept that take interaction into account.

Future requirements for drastic reduction of CO2production and energy consumption will lead to significant changes in the way we see mobility in the years to come. However, the automotive industry has identified significant barriers to the adoption of electric vehicles, including reduced driving range and greatly increased refueling times. Automated cars have the potential to reduce the environmental impact of driving, and increase the safety of motor vehicle travel. The current state-of-the-art in vehicle automation requires a suite of expensive sensors. While the cost of these sensors is decreasing, integrating them into electric cars will increase the price and represent another barrier to adoption. The V-Charge Project, funded by the European Commission, seeks to address these problems simultaneously by developing an electric automated car, outfitted with close-to-market sensors, which is able to automate valet parking and recharging for integration into a future transportation system. The final goal is the demonstration of a fully operational system including automated navigation and parking. This paper presents an overview of the V-Charge system, from the platform setup to the mapping, perception, and planning sub-systems.

In this paper a generic framework for sampling-based partial motion planning along a reference path is presented. The sampling mechanism builds on the specification of a vehicle model and a control law, both of which are freely selectable. Via a closed-loop forward simulation, the vehicle model is regulated onto a carefully chosen set of terminal states aligned with the reference path, generating system-compliant sample trajectories in accordance with the specified system and environmental constraints. The consideration of arbitrary state and input limits make this framework appealing to nonholonomic systems. The rich trajectory set is evaluated in an online sampling-based planning framework, targeting realtime motion planning in dynamic environments. In an example application, a Volkswagen Golf is modeled via a kinodynamic single-track system that is further constrained by steering angle/rate and velocity/acceleration limits. Control is implemented via state-feedback onto piecewise C0-continuous reference paths. Experiments demonstrate the planner's applicability to online operation, its ability to cope with discontinuous reference paths as well as its capability to navigate in a realistic traffic scenario.

This paper addresses decentralized motion planning among a homogeneous set of feedback-controlled, decision-making agents. It introduces the continuous control obstacle ( Cn-CO), which describes the set of Cn-continuous control sequences (and thus trajectories) that lead to a collision between interacting agents. By selecting a feasible trajectory from Cn-CO's complement, a collision-free motion is obtained. The approach represents an extension to the reciprocal velocity obstacle (RVO, ORCA) collision-avoidance methods so that trajectory segments verify Cncontinuity rather than piecewise linearity. This allows the large class of robots capable of tracking Cn-continuous trajectories to employ it for partial motion planning directly-rather than as a mere tool for collision checking. This paper further establishes that both the original velocity obstacle method and several of its recently developed reciprocal extensions (which treat specific robot physiologies only) correspond to particular instances of Cn-CO. In addition to the described extension in trajectory continuity, Cn-CO thus represents a unification of existing RVO theory. Finally, the presented method is validated in simulation-and a parameter study reveals under which environmental and control conditions Cn-CO with admits significantly improved navigation performance compared with inflated approaches based on ORCA.

This paper describes work on multi-robot pattern formation. Arbitrary target patterns are represented with an optimal robot deployment, using a method that is independent of the number of robots. Furthermore, the trajectories are visually appealing in the sense of being smooth, oscillation free, and showing fast convergence. A distributed controller guarantees collision free trajectories while taking into account the kinematics of differentially driven robots. Experimental results are provided for a representative set of patterns, for a swarm of up to ten physical robots, and for fifty virtual robots in simulation.

In this paper we describe a novel and simple to implement yet effective lattice design algorithm, which simultaneously produces input and state-space sampled lattice graphs. The presented method is an extension to the ideas suggested by Bicchi et al. on input lattices and is applicable to systems which can be brought into (2,n) chained form, such as kinematic models of unicycles, bicycles, differential-drive robots and car-like vehicles (pulling several trailers). We further show that a transformation from chained form to path coordinates allows the resulting lattice to be bent along any C1continuous path. We exploit this fact by shaping it along the skeleton of arbitrary structured environments, such as the center of road lanes and corridors. In our experiments in both structured (i.e. on-road) and unstructured (i.e. parking lot) scenarios, we successfully demonstrate for the first time the applicability of lattice-based planning approaches to search queries in arbitrary environments.

In this paper we describe a novel path planning approach for mobile robots operating in indoor environments. In such scenarios, robots must be able to maneuver in crowded spaces, partially filled with static and dynamic obstacles (such as people). Our approach produces smooth, complex maneuvers over large distances through the use of an anytime graph search algorithm applied to a novel multi-resolution state lattice, where the resolution is adapted based on both environmental characteristics and task characteristics. In addition, we present a novel approach for generating fast globally optimal trajectories in constrained spaces (i.e. rooms connected via doors and hallways). This approach exploits offline precomputation to provide extremely efficient online performance and is applicable to a wide range of both indoor and outdoor navigation scenarios. By combining an anytime, multi-resolution lattice-based search algorithm with our precomputation technique, globally optimal trajectories in up to four dimensions (2D position, heading and velocity) are obtained in real-time.

Most of the existing camera calibration toolboxes require the observation of a checkerboard shown by the user at different positions and orientations. This paper presents an algorithm for the automatic detection of checkerboards, described by the position and the arrangement of their corners, in blurred and heavily distorted images. The method can be applied to both perspective and omnidirectional cameras. An existing corner detection method is evaluated and its strengths and shortcomings in detecting corners on blurred and distorted test image sets are analyzed. Starting from the results of this analysis, several improvements are proposed, implemented, and tested. We show that the proposed algorithm is able to consistently identify 80% of the corners on omnidirectional images of as low as VGA resolution and approaches 100% correct corner extraction at higher resolutions, outperforming the existing implementation significantly. The performance of the proposed method is demonstrated on several test image sets of various resolution, distortion, and blur, which are exemplary for different kinds of camera-mirror setups in use.