We present PRVO, a probabilistic variant of Reciprocal Velocity Obstacle (RVO) for decentralized multi-robot navigation under uncertainty. PRVO characterizes the space velocities that would allow each robot to fulfill its share in collision avoidance with specified probability. is modeled as chance constraints over velocity level defined by RVO and takes into account uncertainty associated both state estimation well actuation robot. Since are general computationally intractable, we propose...

Diffusion-based motion planners are becoming popular due to their well-established performance improvements, stemming from sample diversity and the ease of incorporating new constraints directly during inference. However, a primary limitation diffusion process is requirement for substantial number denoising steps, especially when coupled with gradient-based guidance. In this paper, we introduce, in parametric space trajectories, where parameters represented as Bernstein coefficients. We show...

The current paper proposes a trajectory optimization approach for navigating non-holonomic wheeled mobile robot in dynamic environments. obstacle's motion is not known and hence represented by band of predicted trajectories. can account large number obstacle trajectories seeks to avoid each every the sensing range robot. two primary contributions proposed are (1): A computationally efficient method computing intersection space collision avoidance constraints (2): framework connect state...

Standard Model Predictive Control (MPC) or trajectory optimization approaches perform only a local search to solve complex non-convex problem. As result, they cannot capture the multi-modal characteristic of human driving. A global optimizer can be potential solution but is computationally intractable in real-time setting. In this letter, we present MPC capable searching over different driving modalities. Our basic idea simple: run several goal-directed parallel optimizations and score...

The probabilistic velocity obstacle (PVO) extends the concept of (VO) to work in uncertain dynamic environments. In this paper, we show how a robust model predictive control (MPC) with PVO constraints under non-parametric uncertainty can be made computationally tractable. At core our formulation is novel yet simple interpretation MPC as problem matching distribution certain desired distribution. To end, propose two methods. Our first baseline method based on approximating Gaussian Mixture...

In this paper, we present a Model Predictive Control (MPC) framework based on path velocity decomposition paradigm for autonomous driving. The optimization underlying the MPC has two layer structure where in first, an appropriate is computed vehicle followed by computation of optimal forward along it. very nature proposed allows seamless compatibility between layers optimization. A key feature work that it offloads most responsibility collision avoidance to which computationally efficient...

Reactive Collision avoidance for non-holonomic robots is a challenging task because of the restrictions in space achievable velocities. The complexity increases further when multiple are operating tight/cluttered spaces. present paper presents framework specially carved out such situations. But at same time can be easily appended with any existing collision framework. At crux methodology concept non-linear scaling which allows to reactively accelerate/de-accelerate without altering geometric...

We present two real-time trajectory optimizers based on the Cross-Entropy Method for visibility-aware navigation. The approaches differ in handling inequality constraints stemming from bounds motion derivatives, collision avoidance, tracking error, etc. Our first optimizer augments inequalities into cost function, while second one relies a novel GPU accelerated batch projection algorithm. adopt learning-based approach to ensure fast query of occlusion arising environment. Specifically, we...

Computing time optimal motions along specified paths forms an integral part of the solution methodology for many motion planning problems. Conventionally, this control problem is solved considering piece-wise constant parametrization input which leads to convexity and sparsity in optimization structure. However, it also results discontinuous trajectory difficult track. Thus, paper we revisit with primary motivation ensuring a high degree smoothness resulting profile. In particular, solve...

Navigating non-holonomic mobile robots in dynamic environments is challenging because it requires computing at each instant, the space of collision free velocities, characterized by a set highly non-linear and non-convex inequalities. Moreover, uncertainty obstacle trajectories further increases complexity problem, as now becomes imperative to relate velocities confidence measure. In this paper, we present novel perspective towards analyzing solving probabilistic avoidance constraints based...

In this paper, we propose a new model predictive control (MPC) formulation for autonomous driving. The novelty of our MPC stems from the following results. Firstly, adopt an alternating minimization approach wherein linear velocities and angular accelerations are alternately optimized. We show that in contrast to joint optimization, exploits structure problem better, which turn translates reduction computation time. Secondly, explicitly incorporates time dependent non-linear actuator...

Efficient navigation in unknown and dynamic environments is crucial for expanding the application domain of mobile robots. The core challenge stems from non-availability a feasible global path guiding optimization-based local planners. As result, existing planners often get trapped poor minima. In this paper, we present novel optimizer that can explore multiple homotopies to plan high-quality trajectories over long horizons while still being fast enough real-time applications. We build on...

In this paper we introduce a novel framework of generating trajectories which explicitly satisfies the stability constraints such as no-slip and permanent ground contact on uneven terrain. The main contributions are: (1) It derives analytical functions depicting evolution vehicle These functional descriptions enable us to have fast evaluation possible along various directions terrain information is used control shape trajectory. (2) introduces paradigm wherein non-linear time scaling brought...

Railways generally run at a speed of 50–100 Km/hr with lot wind energy being wasted to overcome the drag on vehicle. In this work, design is proposed extract some useful out grazing top surface. A converging air duct passage along spiral casing provided roof which used drive vertical axis turbine. The produced can be an alternator produce electricity. Thus renewable source make railway compartment self-sustainable in requirements.

In this letter, we present a computationally efficient trajectory optimizer that can exploit GPUs to jointly compute trajectories of tens agents in under second. At the heart our is novel reformulation non-convex collision avoidance constraints reduces core computation each iteration large scale, convex, unconstrained Quadratic Program (QP). Importantly, QP structure requires us associated matrix factorization/inverse only once for fixed number agents. Moreover, do it offline and then use...

Autonomous cars and fixed-wing aerial vehicles have the so-called non-holonomic kinematics which non-linearly maps control input to states. As a result, trajectory optimization with such motion model becomes highly non-linear non-convex. In this paper, we improve computational tractability of by reformulating it in terms set bi-convex cost constraint functions along penalty. The part acts as relaxation for while residual penalty dictates how well its output obeys behavior. We adopt an...

Trajectory optimization problems under affine motion model and convex cost function are often solved through the convex-concave procedure (CCP), wherein non-convex collision avoidance constraints replaced with its approximation. Although mathematically rigorous, CCP has some critical limitations. First, it requires a collision-free initial guess of solution trajectory which is difficult to obtain, especially in dynamic environments. Second, at each iteration, involves solving constrained...

This paper proposes a Model Predictive Control (MPC) algorithm for target tracking amongst static and dynamic obstacles. Our main contribution lies in improving the computational tractability reliability of underlying non-convex trajectory optimization. The result is an MPC that runs real-time on laptops embedded hardware devices such as Jetson TX2. approach relies novel reformulations tracking, collision, occlusion constraints induce multi-convex structure resulting We exploit these...