Low-level Planning Research Articles

A Model for Automating the Abstraction of Planning Problems in a Narrative Context

Contemporary automated planning research emphasizes the use of domain knowledge abstractions like heuristics to improve search efficiency. Transformative automated abstraction techniques which decompose or otherwise reformulate the problem have a limited presence, owing to poor performance in key metrics like plan length and time efficiency. In this paper, we argue for a reexamination of these transformative techniques in the context of narrative planning, where classical metrics are less appropriate. We propose a model for automating abstraction by decomposing a planning problem into subproblems which serve as abstract features of the problem. We demonstrate the application of this approach on a low-level problem and discuss key features of the resulting abstract problem. Plans in the abstract problem are shorter, representing summaries of low-level plans, but can be directly translated into low-level plans for the original problem.

SayNav: Grounding Large Language Models for Dynamic Planning to Navigation in New Environments

Semantic reasoning and dynamic planning capabilities are crucial for an autonomous agent to perform complex navigation tasks in unknown environments. It requires a large amount of common-sense knowledge, that humans possess, to succeed in these tasks. We present SayNav, a new approach that leverages human knowledge from Large Language Models (LLMs) for efficient generalization to complex navigation tasks in unknown large-scale environments. SayNav uses a novel grounding mechanism, that incrementally builds a 3D scene graph of the explored environment as inputs to LLMs, for generating feasible and contextually appropriate high-level plans for navigation. The LLM-generated plan is then executed by a pre-trained low-level planner, that treats each planned step as a short-distance point-goal navigation sub-task. SayNav dynamically generates step-by-step instructions during navigation and continuously refines future steps based on newly perceived information. We evaluate SayNav on multi-object navigation (MultiON) task, that requires the agent to utilize a massive amount of human knowledge to efficiently search multiple different objects in an unknown environment. We also introduce a benchmark dataset for MultiON task employing ProcTHOR framework that provides large photo-realistic indoor environments with variety of objects. SayNav achieves state-of-the-art results and even outperforms an oracle based baseline with strong ground-truth assumptions by more than 8% in terms of success rate, highlighting its ability to generate dynamic plans for successfully locating objects in large-scale new environments. The code, benchmark dataset and demonstration videos are accessible at https://www.sri.com/ics/computer-vision/saynav.

Deep Q-Learning for Decentralized Multi-Agent Inspection of a Tumbling Target

As the number of on-orbit satellites increases, the ability to repair or de-orbit them is becoming increasingly important. The implicitly required task of on-orbit inspection is challenging due to coordination of multiple observer satellites, a highly nonlinear environment, a potentially unknown or unpredictable target, and time delays associated with ground-based control. There is a critical need for autonomous, robust, decentralized solutions. To achieve this, we consider a hierarchical, learned approach for the decentralized planning of multi-agent inspection of a tumbling target. Our solution consists of two components: a viewpoint or high-level planner trained using deep reinforcement learning, and a low-level planner that will handle the point-to-point maneuvering of the spacecraft. Operating under limited information, our trained multi-agent high-level policies successfully contextualize information within the global hierarchical environment and are correspondingly able to inspect over 90% of nonconvex tumbling targets, even in the absence of additional agent attitude control.

Hierarchical Control for Head-to-Head Autonomous Racing

We develop a hierarchical controller for head-to-head autonomous racing. We first introduce a formulation of a racing game with realistic safety and fairness rules. A high-level planner approximates the original formulation as a discrete game with simplified state, control, and dynamics to easily encode the complex safety and fairness rules and calculates a series of target waypoints. The low-level controller takes the resulting waypoints as a reference trajectory and computes high-resolution control inputs by solving an alternative approximation formulation with simplified objectives and constraints. We consider two approaches for the low-level planner, constructing two hierarchical controllers. One approach uses multi-agent reinforcement learning (MARL), and the other solves a linear-quadratic Nash game (LQNG) to produce control inputs. The controllers are compared against three baselines: an end-to-end MARL controller, a MARL controller tracking a fixed racing line, and an LQNG controller tracking a fixed racing line. Quantitative results show that the proposed hierarchical methods outperform their respective baseline methods in terms of head-tohead race wins and abiding by the rules. The hierarchical controller using MARL for low-level control consistently outperformed all other methods by winning over 90% of head-to-head races and more consistently adhered to the complex racing rules. Qualitatively, we observe the proposed controllers mimicking actions performed by expert human drivers such as shielding/blocking, overtaking, and long-term planning for delayed advantages. We show that hierarchical planning for game-theoretic reasoning produces competitive behavior even when challenged with complex rules and constraints.

A Hierarchical Manipulation Scheme for Robotic Sorting of Multiwire Cables With Hybrid Vision

A key step in manufacturing multiwire cables in the factory is to sort the wires according to the desired color sequence. Owing to their deformable nature, it is challenging for robots to carry out sorting tasks on wires. Specifically, the wires are very small in size, multiple wires are closely coupled, and the initial color sequence is random. In this article, a new hierarchical manipulation scheme is proposed for the robotic sorting of multiwire cables. The sorting task is carried out by a vision-guided robotic gripper, whose control objective is formulated in two layers: 1) the high-level planner outputs an optimal grasping sequence and the grasping position, according to the initial status of wires; and 2) the low-level planner shapes the gripper's reference trajectory by monitoring the wires' changes online. The proposed hierarchical manipulation scheme explores the advantages of both 3-D and 2-D vision to effectively deal with the existing issues (e.g., overlapping and occlusion) of multiple wires and also unforeseen changes during the sorting task (e.g., slipping and collision), which then guarantees the feasibility of wire sorting. Experimental results in various scenarios are presented to validate the performance of the proposed scheme.

A Trajectory Optimization-Based Intersection Coordination Framework for Cooperative Autonomous Vehicles

Since vehicles from multiple roads frequently merge at intersections, it formulates a typical traffic bottleneck of modern transportation systems. Proper vehicle coordination and motion plan at road intersections are of importance to guarantee safety as well as improving the traffic throughput, fuel efficiency and so on. In this paper, we try to present a general dedicated intersection coordination framework for autonomous vehicles, where both high- and low-level planners are appropriately designed and integrated. In the high-level planner, two different strategies are formulated to coordinate the autonomous vehicles to generate <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">reference trajectories</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">feasible “tunnels</i> ”, respectively. Especially, a novel space-time-block based resource allocation scheme is presented to describe the feasible tunnels. Furthermore, to avoid collisions with unexpected obstacles such as pedestrians, bicycles or other vehicles with human drivers, a low-level planner is designed to generate practical trajectories based on the solutions from the high-level planner, according to their local on-board observations. Simulations and practical experiments are carried out, to show that our proposed coordination framework can achieve obvious performance advantages in various traffic metrics, including the throughput, fairness in driving maneuvers and driving comfort, etc. We also find that the high-level planner is effective in eliminating possible <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">deadlocks</i> among autonomous vehicles, which is rarely discussed in existing investigations.

Global-Position Tracking Control for Three-Dimensional Bipedal Robots Via Virtual Constraint Design and Multiple Lyapunov Analysis

Abstract A safety-critical measure of legged locomotion performance is a robot's ability to track its desired time-varying position trajectory in an environment, which is herein termed as “global-position tracking.” This paper introduces a nonlinear control approach that achieves asymptotic global-position tracking for three-dimensional (3D) bipedal robots. Designing a global-position tracking controller presents a challenging problem due to the complex hybrid robot model and the time-varying desired global-position trajectory. Toward tackling this problem, the first main contribution is the construction of impact invariance to ensure all desired trajectories respect the foot-landing impact dynamics, which is a necessary condition for realizing asymptotic tracking of hybrid walking systems. Thanks to their independence of the desired global position, these conditions can be exploited to decouple the higher-level planning of the global position and the lower-level planning of the remaining trajectories, thereby greatly alleviating the computational burden of motion planning. The second main contribution is the Lyapunov-based stability analysis of the hybrid closed-loop system, which produces sufficient conditions to guide the controller design for achieving asymptotic global-position tracking during fully actuated walking. Simulations and experiments on a 3D bipedal robot with twenty revolute joints confirm the validity of the proposed control approach in guaranteeing accurate tracking.

Effectively Incorporating Weighted Cost-to-go Heuristic in Suboptimal CBS (Extended Abstract)

Conflict-Based Search (CBS) is a popular multi-agent path finding (MAPF) solver that employs a low-level single agent planner and a high-level constraint tree to resolve conflicts. The majority of modern MAPF solvers focus on improving CBS by reducing the size of this tree through various strategies with few methods modifying the low level planner. All low level planners in existing CBS methods use an unweighted cost-to-go heuristic, with suboptimal CBS methods also using a conflict heuristic to help the high level search. Contrary to prevailing beliefs, we show that the cost-to-go heuristic can be used significantly more effectively by weighting it in a specific manner alongside the conflict heuristic. We introduce two variants of doing so and demonstrate that this change can lead to 2-100x speedups in certain scenarios. Additionally, we show the first theoretical relation of prioritized planning and bounded suboptimal CBS and demonstrate that our methods are their natural generalization.

Multi-Agent Path Finding with Temporal Jump Point Search

Temporal Jump Point Search (JPST) is a recently introduced algorithm for grid-optimal pathfinding among dynamic temporal obstacles. In this work we consider JPST as a low-level planner in Multi-Agent Path Finding (MAPF). We investigate how the canonical ordering of JPST can negatively impact MAPF performance and we consider several strategies which allow us to overcome these limitations. Experiments show our new CBS/JPST approach can substantially improve on CBS/SIPP, a contemporary and leading method from the area.

Dynamic Causal Modelling of Hierarchical Planning

Hierarchical planning (HP) is a strategy that optimizes the planning by storing the steps towards the goal (lower-level planning) into subgoals (higher-level planning). In the framework of model-based reinforcement learning, HP requires the computation through the transition value between higher-level hierarchies. Previous study identified the dmPFC, PMC and SPL were involved in the computation process of HP respectively. However, it is still unclear about how these regions interaction with each other to support the computation in HP, which could deepen our understanding about the implementation of plan algorithm in hierarchical environment. To address this question, we conducted an fMRI experiment using a virtual subway navigation task. We identified the activity of the dmPFC, premotor cortex (PMC) and superior parietal lobe (SPL) with general linear model (GLM) in HP. Then, Dynamic Causal Modelling (DCM) was performed to quantify the influence of the higher- and lower-planning on the connectivity between the brain areas identified by the GLM. The strongest modulation effect of the higher-level planning was found on the dmPFC→right PMC connection. Furthermore, using Parametric Empirical Bayes (PEB), we found the modulation of higher-level planning on the dmPFC→right PMC and right PMC→SPL connections could explain the individual difference of the response time. We conclude that the dmPFC-related connectivity takes the response to the higher-level planning, while the PMC acts as the bridge between the higher-level planning to behavior outcome.

Motion Intention Prediction and Joint Trajectories Generation Toward Lower Limb Prostheses Using EMG and IMU Signals

Powered intelligent lower limb prostheses have been gaining interest as they provide functionality for walking on different terrains. This study proposes a hierarchical planner for intelligent lower limb prostheses based on sensor fusion and a central pattern generator (CPG). Electromyographic (EMG) and inertial measurement unit (IMU) signals were recorded and fused in the feature and decision levels. The high-level planner consists of cascade classifiers with a gait phase dependence. A secondary classifier for each stand phase and swing phase was developed to recognize five gait patterns. The mid-level planner was designed as a walking frequency estimator based on the Newton method. The low-level planner incorporates the CPG models, achieving the planning of lower limb joint trajectories. The proposed layered planner can recognize the users&#x2019; walking intention and gait speed in real-time, ensuring coordination between the joints of both legs. Eight healthy subjects were recruited, and the average accuracy of motion pattern recognition reached 99.13&#x0025; and 99.39&#x0025; for the standing and swing phases, respectively. The relative root mean square error (RMSE) of the walking frequency estimate was 2.3&#x0025; under approximately 3.5 m/s gait speed. The promising results indicate that this method effectively predicts the continuous joint angle for lower limb prostheses.

Optimal Control of Connected and Automated Vehicles at Multiple Adjacent Intersections

In this paper, we establish a decentralized optimal control framework for connected and automated vehicles (CAVs) crossing multiple adjacent, multi-lane signal-free intersections to minimize energy consumption and improve traffic throughput. Our framework consists of two layers of planning. In the upper-level planning, each CAV computes its optimal arrival time at each intersection recursively along with the optimal lane to improve the traffic throughput. In the low-level planning, we formulate an energy-optimal control problem with interior-point constraints, the solution of which yields the optimal control input (acceleration/deceleration) of each CAV to cross the intersections at the time specified by the upper-level planning. Moreover, we extend the results of the proposed bi-level framework to include a bounded steady-state error in tracking the optimal position of the CAVs. Finally, we demonstrate the effectiveness of the proposed framework through simulation for symmetric and asymmetric intersections and comparison with traditional signalized intersections.

Multi-objective rescue path optimization for passenger ship accident under tilt

In order to research the rescue path problem in the accident of passenger ships under tilt, this paper establishes a multi-objective rescue path optimization model under tilt effect. By analyzing the fuzzy time and fuzzy risk, the objective functions of this model are optimal satisfaction function and optimal risk function. Related constraints are also described mathematically. The PSO-GA (particle swarm and genetic) hybrid algorithm is used to solve the model when designing the algorithm. Two-level planning is incorporated in the algorithm, the best solution in the lower-level planning is assigned to the upper-level, and the upper-level plan feeds back the result to the lower level, and finally the global optimal Pareto solution is obtained. Decision makers can choose appropriate solutions based on their preference. The simulation experiment compares the multi-objective rescue path optimization model with the traditional time-optimal model. Among the three optimal solution sets, solution 1 decreases by 3.36% in risk and the satisfaction rate increases by 69.44%. Solution 2 rose by 13.96% in risk, but the satisfaction increased by 87.93%, and the risk of solution 3 decreased by 11.41%, while the satisfaction increased by 52.41%. The results show that the established model is reasonable and the algorithm is feasible.

Multi-Robot Robust Motion Planning based on Model Predictive Priority Contouring Control with Double-Layer Corridors

Disturbance poses a major challenge for the safety and real-time performance of robust robot motion planning. To address the disturbance while improving the real-time performance of multi-robot robust motion planning, a model predictive priority contouring control method is proposed. First, an improved conflict-based search (ICBS) planner is utilized to plan reference paths. The low-level planner of the conflicted-based search (CBS) planner is replaced by the hybrid A* planner and reference paths are adopted as an initial guess of model predictive priority contouring control. Second, double-layer corridors are proposed to provide safety guarantees, which include static-layer corridors and dynamic-layer corridors. The static-layer corridors are generated based on reference paths and the dynamic-layer corridors are generated based on the relative positions and velocities of robots. The double-layer corridors are applied as safety constraints of model predictive priority contouring control. Third, a prioritization mechanism is devised to improve computational efficiency. Priorities are assigned according to each robot’s task completion percentage. Based on the assigned priority, multiple robots are grouped, and each group executes the model predictive priority contouring control algorithm to acquire trajectories. Finally, our method is compared with the centralized method and the soft constraint-based DMPC. Simulations verify the effectiveness and real-time performance of our approach.

Chance-Constrained Multilayered Sampling-Based Path Planning for Temporal Logic-Based Missions

This article introduces a robust and safe path planning algorithm in order to satisfy mission requirements specified in linear temporal logic (LTL). When a path is planned to accomplish a mission, it is possible for a robot to fail to complete the mission or collide with obstacles due to noises and disturbances in the system. Hence, we need to find a robust path against possible disturbances. We introduce a robust path planning algorithm, which maximizes the probability of success in accomplishing a given mission by considering disturbances, while minimizing the moving distance of a robot. The proposed method can guarantee the safety of the planned trajectory by incorporating an LTL formula and chance constraints in a hierarchical manner. A high-level planner generates a discrete plan satisfying the mission requirements specified in LTL. A low-level planner builds a sampling-based rapidly exploring random tree search tree to minimize both the mission failure probability and the moving distance while guaranteeing the probability of collision with obstacles to be below a specified threshold. We have analyzed properties of the proposed algorithm theoretically and validated the robustness and safety of paths generated by the algorithm in simulation and experiments using a quadrotor.

ECBS with Flex Distribution for Bounded-Suboptimal Multi-Agent Path Finding

Multi-Agent Path Finding (MAPF) is the problem of finding collision-free paths for multiple agents. CBS is a leading optimal two-level MAPF solver whose low level plans optimal paths for single agents and whose high level runs a best-first search on a Constraint Tree (CT) to resolve the collisions between the paths. ECBS, a bounded-suboptimal variant of CBS, speeds up CBS by reducing the number of collisions that need to be resolved on the high level. It achieves this by generating bounded-suboptimal paths with fewer collisions with the paths of the other agents on the low level and expanding bounded-suboptimal CT nodes that contain fewer collisions on the high level. In this paper, we propose Flexible ECBS (FECBS) that further reduces the number of collisions that need to be resolved on the high level by using looser suboptimal bounds on the low level while still providing bounded-suboptimal solutions. Instead of requiring the cost of each path to be bounded-suboptimal, FECBS requires only the overall cost of the paths to be bounded-suboptimal, which gives us the freedom to distribute the cost leeway among different agents according to their needs. Our empirical results show that FECBS can solve more MAPF instances than state-of-the-art ECBS variants within 5 minutes.

A Bi-Level Model for Green Freight Transportation Pricing Strategy Considering Enterprise Profit and Carbon Emissions

Due to the impact of COVID-19, enterprises need effective pricing strategies to improve profits and viability. In order to fill research gaps in the literature relating to market competition among different freight modes and to adjust the freight transportation structure by optimizing transportation prices, we propose a multi-objective bi-level programming pricing model that considers market competition and the carbon emissions of the freight system in China. First, an objective function in upper-level planning was used to improve logistics enterprise profits and reduce the total carbon emissions of the freight system. Then, a generalized cost function for the freight transportation mode was designed to quantify the market competition among different transportation modes, and a user equilibrium assignment model was established to obtain the results of cargo flow assignment in lower-level planning. To solve the model, a sensitivity analysis algorithm was designed, and a logistics network example was used to verify the effectiveness of the model. The experimental results show that reasonable freight price adjustment can effectively increase enterprise profits and reduce the total carbon emissions of the freight system. In this paper, we provide a new method for freight pricing research, considering the market competition of multiple transportation modes, and provide a new idea for adjusting the transportation structure through freight price optimization, which will play a positive role in promoting the development of green freight.

Coordinated Planning of Distribution System With RES, DSTATCOM, and Protective Devices

In the competitive energy market, ensuring a reliable power supply becomes a major concern and accordingly, the distribution system must be equipped with protective devices to reduce frequency and duration of fault. In addition, large-scale integration of renewable energy sources (RES) is required because of the enormous load growth. To alleviate the challenges of large-scale RES incorporation, Distribution Static Compensator (DSTATCOM) must be incorporated along with RES. In case of any fault, these devices can improve overall reliability by continuing power supply to the isolated loads; thus, more efficient planning can be achieved with the concurrent utilization of RES, DSTATCOM, and protective devices. Accordingly, this study introduces a bilevel planning model where the optimal integration of RES and DSTATCOM is determined in upper level planning (ULP) and the protective devices are incorporated in lower level planning (LLP). In the bilevel model, for each ULP population, the LLP model is executed and the optimal results are feedback to the ULP to calculate ULP objectives. Therefore, all the ULP populations are combined with the optimal results of LLP. Moreover, to reflect the reduction of available budget due to various cost-cutting efforts of a distribution company, an independent RES owner is introduced to invest in RES and proper coordination of both parties is ensured with a long-term contract price. Furthermore, a computationally efficient 2m point estimation method is applied to capture the uncertainty of RES, load, and energy price. The proposed planning model is evaluated in 119-bus test system and the simulated results are presented and analyzed in detail.

Autonomous Driving at Intersections: A Behavior-Oriented Critical-Turning-Point Approach for Decision Making

Left turning at unsignalized intersectionis one of the most challenging tasks for urban automated driving, due to the various shapes of intersections and rapidly changing nature of the driving scenarios. This article addresses the challenges of decision making at highly uncertain intersections of different shapes by proposing a generalized critical turning point (CTP)-based hierarchical decision-making and planning method. The high-level planner takes the road map information and generates CTPs using a parameterized extraction model that is proposed and verified by naturalistic driving data. The CTPs are used to generate behavior-oriented paths that could be adapted to various intersections. These modifications help to assure high searching efficiency of the planning process. The low-level planner makes real-time, 2-D planning using a partially observable Markov decision process solver, which can handle the uncertainties of the intersections and make less conservative yet safe actions. With proper modifications, our proposed method can make commute-efficient 2-D planning decisions at unsignalized intersections of various shapes in real time.

Evolutionary bi-level model for optimizing ticket fares and operations profit of Taiwan high-speed rail

The Taiwan High Speed Rail (THSR) has transformed transportation in western Taiwan. Increasing costs of construction and operation have rendered THSR tickets much more expensive than other forms of transportation. The effects of ticket price benefits on the relevant transportation agency and the distribution of passenger flow have been analyzed to improve the profitability of the THSR. However, relying on analysis of passenger flow information alone may be insufficient to the adjustment of ticket fares. An evolutionary optimization model for maximizing the profits of the transportation agency that considers conflicts between executive decision making and passenger flow was developed herein. To resolve these conflicts, the bi-level planning approach was applied to consider both upper-level and lower-level planning and reflect passenger behavior and transportation profit management. Therefore, developing business plans and strategies that maximize cost efficiency while optimally balancing passenger satisfaction and profitability could improve THSR performance. Results indicated that optimal fare rates should be divided into three groups: 3.1, 3.0, and 4.2 NTD/person-km for long, medium, and short distances, respectively. This tool could be used to model and validate THSR fare adjustments and as a reference for authorities when making policy recommendations.