Abstract: An alternate landing site selector system and apparatus utilizing a hierarchy calculator for automatically ranking a plurality of alternate landing sites by assigning a suitability value to each of the alternate landing sites corresponding to that site's suitability for landing, wherein the assigned suitability value is based on at least distance from the aircraft, runway length, terrain, or weather conditions. The system may assist the pilot (crew) in resolving or reducing the urgency of an emergency or in executing a proposed solution to a given emergency, or may direct the aircraft autopilot in executing the suggested (and selected) solution. In either case, the system of the present invention provides the pilot with time to consider an emergency and its resolution (while the aircraft is directed toward a safe configuration or flight condition.

Abstract: Methods and system for minimizing soil compaction relating to operations performed by agricultural machines. The agricultural machines can be guided so that at least one engagement body travels along an existing track path that was established by a prior travel of the same or different agricultural machine. Further, travel plans can determine the sequence in which the agricultural machine(s) travel in areas in the field so that travel on the track path(s) can be initiated in drier regions in the field. Operation of agricultural machines can also be adjusted to minimize deviations of other agricultural machines from existing track paths. For example, a windrower can operate a diverter to deposit a windrow at an offset location, and with a non-linear pattern, that may limit a baler from deviating from an existing track path while collecting crop material from the windrow, while also assisting in forming uniformly shaped crop bales.

Abstract: A control device includes a control unit that executes automatic parking control for parking the moving body at the parking position based on recognition data of an external environment of a moving body and the parking position and registers the parking position as a designated parking position. The control unit executes first registration of registering the designated parking position based on first recognition data and executes second registration of registering the designated parking position based on second recognition data. The control unit continues the second registration without executing automatic movement control when execution of the automatic movement control is instructed during execution of the second registration in new registration of the designated parking position.

Abstract: A method of autonomous vehicle control, comprising: receiving an image of a lenticular human-imperceptible marker embedded in an element of an environment that an autonomous vehicle is moving in, the marker having a pattern usable for determining positional data of the moving vehicle, the image captured using human-invisible light, analyzing the received image of the human-imperceptible marker, and controlling the autonomous vehicle based on the analyzed image of the human-imperceptible marker.

Abstract: A parking assistance device includes: a vehicle location acquiring unit for acquiring the current location of a vehicle; an object information acquiring unit for acquiring object information that includes information relating to locations of objects in the vicinity of the vehicle; and a target route setting unit for setting a plurality of routes between the current location and a predetermined parking space, identifying the route, of the plurality of routes, with the longest distance to the object, and, based on the identified route, setting a target route for the vehicle to travel from the current location to the predetermined parking space.

Abstract: A method for creating a guidance trajectory for a first transportation vehicle which includes receiving swarm data for a route section that represent at least one motion trajectory of at least one second transportation vehicle traveling through this route section; determining a distance from the at least one motion trajectory represented by the swarm data to a determined roadway edge for the route section; and using an electronic computing device to apply the at least one motion trajectory to ascertain the guidance trajectory for the first transportation vehicle in response to the distance associated with the respective motion trajectory being greater than or equal to a predefined limit value for the route section, and to provide the guidance trajectory for the first transportation vehicle, as a result of which the first transportation vehicle is transversely controlled using the guidance trajectory.

Abstract: The present disclosure relates to a vehicle monitoring and control system and a method for controlling at least one function of a vehicle. For example, a Recreational Vehicle (RV) monitoring and control system and a method for controlling at least one function of a RV are proposed. The vehicle monitoring and control system comprises at least one controller. The at least one controller is configured to receive sensor input data from at least two different types of sensor units. The at least one controller is configured to process the received sensor input data into a combined sensor input data and to compare the combined sensor input data against at least one threshold. Based on a result of the comparison, the at least one controller is configured to control at least one function of the vehicle.

Abstract: Autonomous excavator has developed rapidly in recent years because of a shortage of labor and hazardous working environments for operating excavators. Presented herein are embodiments of a novel hierarchical planning system for autonomous machines, such as excavators. In one or more embodiments, the overall planning system comprises a high-level task planner for task division and base movement planning, and general sub-task planners with motion primitives, which include both arm and base movement in the case of an excavator. Using embodiments of the system architecture, experiments were performed for the trench and pile removal tasks in the real world and for the large-scale material loading tasks in a simulation environment. The results show that the system architecture embodiments and planner method embodiments generate effective task and motion plans that perform well in autonomous excavation.

Abstract: A system can use semantic images, lidar images, and/or 3D bounding boxes to determine mobility parameters for objects in the semantic image. In some cases, the system can generate virtual points for an object in a semantic image and associate the virtual points with lidar points to form denser point clouds for the object. The denser point clouds can be used to estimate the mobility parameters for the object. In certain cases, the system can use semantic images, lidar images, and/or 3D bounding boxes to determine an object sequence for an object. The object sequence can indicate a location of the particular object at different times. The system can use the object sequence to estimate the mobility parameters for the object.

Abstract: A materials handling vehicle includes a camera, odometry module, processor, and drive mechanism. The camera captures images of an identifier for a racking system aisle and a rack leg portion in the aisle. The processor uses the identifier to generate information indicative of an initial rack leg position and rack leg spacing in the aisle, generate an initial vehicle position using the initial rack leg position, generate a vehicle odometry-based position using odometry data and the initial vehicle position, detect a subsequent rack leg using a captured image, correlate the detected subsequent rack leg with an expected vehicle position using rack leg spacing, generate an odometry error signal based on a difference between the positions, and update the vehicle odometry-based position using the odometry error signal and/or generated mast sway compensation to use for end of aisle protection and/or in/out of aisle localization.

Abstract: A vehicle control device is disclosed. The vehicle control device controls a vehicle on the basis of one or more detection times each representing a distance measured by a distance measuring device mounted on the vehicle. The distance measuring device measures a distance to an object around the vehicle by transmitting and receiving sound waves. The vehicle control device includes acquisition circuitry and estimated coordinate calculation circuitry. The acquisition circuitry acquires the one or more detection times. The estimated coordinate calculation circuitry calculates estimated coordinates indicating an estimated position of an object in a case where the object is present in a neighborhood of the distance measuring device but a distance to the object is not measured. The estimated coordinates is calculated on the basis of the one or more detection times having been acquired by the acquisition circuitry.

Abstract: A method and a device for vehicle control are disclosed. The method includes: in response to a driving mode switching instruction, detecting whether an electric quantity of a low-voltage battery of the vehicle is higher than a preset threshold value, the preset threshold value being determined according to an electric quantity value required for the vehicle to park; and when the electric quantity of the low-voltage battery is higher than the preset threshold value, controlling the vehicle to enter a target driving mode corresponding to the driving mode switching instruction.

Abstract: One embodiment of a method for controlling a robot includes receiving sensor data associated with an environment that includes an object; applying a machine learning model to a portion of the sensor data associated with the object and one or more trajectories of motion of the robot to determine one or more path lengths of the one or more trajectories; generating a new trajectory of motion of the robot based on the one or more trajectories and the one or more path lengths; and causing the robot to perform one or more movements based on the new trajectory.

Abstract: Methods and apparatus for platoon behaviors that discourage other vehicles from interfering.

Abstract: The present disclosure relates to an autonomous vehicle and a remote control request method therefor. An exemplary embodiment of the present disclosure provides an autonomous vehicle including: a processor configured to determine whether the autonomous vehicle is in a state in which remote control is enabled to be performed when the remote control of the autonomous vehicle is required, and to request a control system to perform the remote control when the autonomous vehicle is in the state in which the remote control is enabled to be performed; and a communication device configured to perform communication with the control system under control of the processor.

Abstract: Embodiments of the disclosure provide a method and apparatus for trajectory prediction, a device and a storage medium. The method includes that: location information of a reference end point of a moving object is determined according to location information of the moving object; a candidate trajectory set including multiple candidate trajectories is determined according to the location information of the moving object and the location information of the reference end point, wherein location information of an end point of each candidate trajectory is different from the location information of the reference end point; and a target trajectory of the moving object is determined from the candidate trajectory set.

Abstract: The present invention relates to an augmented reality system (AR system) for teaching a user of an application. The AR system comprises a detection unit that detects a real environment and provides it as environmental data, wherein at least one work unit is arranged in the environment; a communication unit that receives process data from the work unit; a control unit that processes the process data and environmental data, wherein a virtual image can be created based on the process data; and a display unit that displays the real environment and the virtual image in relation to one another to a user. The AR system further comprises an audio output unit that, on the basis of the process data, outputs audio signals for instructing a user, wherein the audio signals are preferably matched to the virtual image.

Abstract: Provided is a navigation system for a leader vehicle leading follower vehicles. The leader vehicle is configured to transmit real-time movement data to follower vehicles. The follower vehicles each include a signal receiver for receiving the data from the leader vehicle and a processor configured to determine a set of active maneuvering instructions for the respective follower vehicle based on at least a portion of the data received from the leader vehicle and actuate the respective follower vehicle to execute the set of active maneuvering instructions.

Abstract: Multi-media parcel delivery systems and associated methods are provided herein including systems capable of delivering a parcel through air and one or more bodies of water. In certain embodiments, an aerial vehicle is configured to couple to the parcel using a cable. A control system is used to control operating parameters including velocity, altitude, and pose of the aerial vehicle, and a length and orientation of the cable extending between the aerial vehicle and the parcel.

Abstract: Techniques for traversing in an environment that includes at least one obstacle, by a mobile autonomous system, to a destination in the environment, are presented. The techniques can include generating, prior to the mobile autonomous system commencing activity in the environment, a graph including a plurality of vertices representing positions in the environment and a plurality of edges between vertices representing feasible transitions by the mobile autonomous vehicle in the environment; annotating the graph with at least one edge connecting a representation of a present position of the mobile autonomous system to a vertex of the graph; determining, based on the graph, a path from the present position of the mobile autonomous system in the environment to the destination; and traversing the environment to the destination, by the mobile autonomous system, based on the path.

Abstract: A method of soliciting an occupant intervention for a simultaneous localization and mapping-based navigation system. A vehicle path plan and a motion control instruction are created and a suspected obstacle in a drive path is identified, wherein the drive path provides a portion of the vehicle path plan. The method further includes providing an indication on a first display that an intervention by an occupant is requested and determining if the occupant is paying attention to the indication on the first display. A system of performing the method includes a controller configured to execute instructions executing the method.

Abstract: An automated valet parking system, an automated valet parking method, and an automated valet parking infrastructure, and a vehicle having an automated valet parking feature are disclosed. In particular, the vehicle can autonomously move to and park in a designated parking spot by communicating with the infrastructure. In addition, the vehicle can autonomously move to a pickup area from a parking spot by communicating with the infrastructure.

Abstract: The present disclosure is directed to deviating from a planned path for an autonomous vehicle. In particular, a computing system comprising one or more computing devices physically located onboard an autonomous vehicle can identify one or more boundaries at least in part defining a lane in which the autonomous vehicle is traveling along a path of a planned route. Responsive to identifying one or more obstructions ahead of the autonomous vehicle along the path, the computing system can: determine one or more deviations from the path that would result in the autonomous vehicle avoiding the obstruction(s) and at least partially crossing at least one of the one or more boundaries; and generate, based at least in part on the deviation(s), a motion plan instructing the autonomous vehicle to deviate from the path such that it avoids the obstruction(s) and continues traveling along the planned route.

Abstract: A network computing system can coordinate on-demand transport serviced by transport providers operating throughout a transport service region. The transport providers can comprise a set of internal autonomous vehicles (AVs) and a set of third-party AVs. The system can receive a transport request from a requesting user of the transport service region, where the transport request indicates a pick-up location and a destination. The system can determine a subset of the transport providers to service the respective transport request, and executing a selection process among the subset of the transport providers to select a transport provider to service the transport request. The system may then transmit a transport assignment to the selected transport provider to cause the selected transport provider to service the transport request.

Abstract: A method for autonomously servicing a first cleaning component of a battery-operated mobile device, including: inferring, with a processor of the mobile device, a value of at least one environmental characteristic based on sensor data captured by a sensor disposed on the mobile device; actuating, with a controller of the mobile device, a first actuator interacting with the first cleaning component to at least one of: turn on, turn off, reverse direction, and increase or decrease in speed such that the first cleaning component engages or disengages based on the value of at least one environmental characteristic or at least one user input received by an application of a smartphone paired with the mobile device; and dispensing, by a maintenance station, water from a clean water container of the maintenance station for washing the first cleaning component when the mobile device is docked at the maintenance station.

Abstract: A method for autonomously planning work duties of a robot within an environment of the robot, including: obtaining, with a processor of the robot, first data indicative of a presence or an absence of at least one human within the environment at a particular time; actuating, with the processor, the robot to execute work duties based on the first data, wherein: the robot executes the work duties when the first data indicates the absence of the at least one human from the environment; and the work duties comprise cleaning at least a portion of the environment; capturing, with at least one sensor of a plurality of sensors disposed on the robot, second data while the robot executes the work duties; and altering, with the processor, a navigational route of the robot or the work duties based on the second data.

Abstract: An automatic working system that includes a self-moving device and a positioning device. The self-moving device includes a movement module, a task execution module. The positioning device is configured to detect a current position of the self-moving device. The automatic working system includes: a storage unit, configured to store a working area map, and: a map confirmation procedure, which including: providing a drive circuit instruction to move along a working area boundary, and receiving a confirmation signal from a user to complete the map confirmation procedure; and a working procedure including providing a drive circuit instruction to move within a working area defined by the map and execute the working task; and a control module, configured to monitor an output of the positioning device to execute the map confirmation procedure and execute the working procedure after the map confirmation procedure is completed.

Abstract: A robot cleaner and a control method thereof are disclosed. A robot cleaner, according to an embodiment of the present invention, comprises: a tilt sensor module; a distance sensor module; and a light amount sensor module. Each piece of sensed information is transmitted to a lifting information calculation module. The lifting information calculation module calculates information on whether the robot cleaner is lifted, by using each piece of the transmitted information. If it is calculated that the robot cleaner is lifted off a floor, a power module is stopped. In addition, if it is calculated that the robot cleaner is not lifted off the floor, the power module is driven. Accordingly, when a user lifts the robot cleaner, injuries to the user caused by operation of a drive module can be prevented.

Abstract: Operating an autonomous grounds maintenance machine includes determining a travel path for the machine to reach a destination waypoint in a zone of a work region. The travel path is analyzed for problem areas based on a predetermined terrain map. Rotations of the machine may be planned even before the machine begins to travel down the path to proactively prevent the machine from becoming trapped.

Abstract: An apparatus for generating a lane network includes a processor configured to calculate, for each of individual points in a predetermined region connecting an entry lane and an exit lane, a probability distribution indicating a probability that the point lies on a standard trajectory, based on trajectories of one or more vehicles that traveled through the region, generate two virtual borders each connecting the positions of respective edges of the entry lane and the exit lane, set, of the individual points in the predetermined region, at least two points that lie between the two virtual borders and at which the variance of the probability distribution is not greater than a predetermined variance threshold, as control points, and generate a Bézier curve, based on the at least two control points, as a line connecting two lanes in the predetermined region in a lane network.

Abstract: In various examples, an end-to-end perception evaluation system for autonomous and semi-autonomous machine applications may be implemented to evaluate how the accuracy or precision of outputs of machine learning models—such as deep neural networks (DNNs)—impact downstream performance of the machine when relied upon. For example, decisions computed by the system using ground truth output types may be compared to decisions computed by the system using the perception outputs. As a result, discrepancies in downstream decision making of the system between the ground truth information and the perception information may be evaluated to either aid in updating or retraining of the machine learning model or aid in generating more accurate or precise ground truth information.

Abstract: The present invention relates to an autonomous omnidirectional drive unit including a chassis (40) defining a front third (41), which contains a first drive wheel (11) with a first gearbox (13), a second drive wheel (21) with a second gearbox (23) coaxial with a horizontal geometric axis (EH), and a battery (30) between same, an intermediate third (42) which contains a first motor (12) and a second motor (22) connected to the first and second drive wheels (11, 21) and a vertical housing (70) between same concentric with a vertical geometric axis (EV); and a rear third which contains at least one caster wheel (50), the first and second motors (12, 22) being controlled by a control device (80) which generates control commands considering the distance (D) existing between the horizontal and vertical geometric axes (EH, EV).

Abstract: Provided are an information processing device and a mobile robot capable of ensuring the accuracy of estimation of a self-position even with an environmental change and maintaining the consistency of the self-position on a map. The information processing device of the mobile robot includes a control section and a detection section configured to detect a distance to a peripheral object and the direction thereof as detection information. The control section includes a storage, a map producer configured to produce a peripheral map, an existing map self-position estimator configured to estimate a current self-position based on an existing map, a current map self-position estimator configured to estimate the current self-position based on a current map, a reliability evaluator configured to evaluate the reliability of each of the estimated self-positions, and a self-position updater configured to update either one of the self-positions based on the reliability.

Abstract: A method for preprocessing a set of feasible transfers within a multimodal transportation network of predetermined stations, comprising, for each trip in the multimodal transportation network hereafter called origin trip: (a) for each station (pti) of the origin trip (t), computing at this station (pti) an earliest arrival/change time associated with all transportation modes (m) of the multimodal transportation network; (b) for at least one transfer of the set of feasible transfers from a station (pti) on the origin trip (t) to a reachable station (puj) on a target trip (u), computing, at each station (puk>j) of the target trip (u) after the reachable station (puj), a value of the earliest arrival/change time specifically associated with the transportation mode (mu) of the multimodal transportation network used by the target trip (u); (c) removing the transfer only if determining that each computed value of the earliest arrival/change time is not improved by the transfer; (d) outputting the set of feasible

Abstract: A system and method for providing long term and key intentions for trajectory prediction that include receiving image data and LiDAR data associated with RGB images and LiDAR point clouds that are associated with a surrounding environment of an ego agent and processing a long term and key intentions for trajectory prediction dataset (LOKI dataset) that is utilized to complete joint trajectory and intention prediction for heterogeneous traffic agents. The system and method also include encoding a past observation history of each of the heterogeneous traffic agents and sampling a respective goal. The system and method further include decoding and predicting future trajectories associated with each of the heterogeneous traffic agents based on data included within the LOKI dataset, the encoded past observation history, and the respective goal.

Abstract: System, methods, and computer-readable media for an object path prediction model, and an associated training technique, to output a path that is considered an object collision path. The object path prediction model outputs a set of predicted paths for the object that are outputted to a trained planning algorithm, which includes paths that are most likely to occur and a path that is considered an object collision path. The predicted paths are sent to the trained planning algorithm and used for planning a trajectory for the autonomous vehicle that is associated with a low probability of colliding with or taking a sudden evasive action to avoid the object.

Abstract: A method and an apparatus for generating sensor data for controlling an autonomous vehicle in an environment is provided, such as driverless transport vehicles in a factory for example. Sensor positions of static sensors and the sensors of autonomous vehicles are defined in a global coordinate system on the basis of an environment model, such as a BIM model for example. Sensor data is centrally generated in this global coordinate system for all sensors as a function of these sensor positions. The sensor data is then transformed into a local coordinate system of an autonomous vehicle and transferred for controlling the autonomous vehicle.

Abstract: A platooning control apparatus may include: a navigation unit configured to guide an ego vehicle to a destination set by a driver; a driving module configured to drive the ego vehicle; and a control unit configured to primarily select platooning groups based on the destination set in the navigation unit, analyze platooning information of the primarily selected platooning groups, finally decide any one of the primarily selected platooning groups, and then control the driving module to join the finally decided platooning group.

Abstract: A system for context-aware decision making of an autonomous agent includes a computing system having a context selector and a map. A method for context-aware decision making of an autonomous agent includes receiving a set of inputs, determining a context associated with an autonomous agent based on the set of inputs, and optionally any or all of: labeling a map; selecting a learning module (context-specific learning module) based on the context; defining an action space based on the learning module; selecting an action from the action space; planning a trajectory based on the action S260; and/or any other suitable processes.

Abstract: Techniques are disclosed for utilizing synchronized robot observations. The techniques may include receiving first log data associated with a first vehicle, receiving second log data associated with a second vehicle, determining that a first portion of the first log data is time related to a second portion of the second log data, and using the first portion and the second portion for mapping and/or training models.

Abstract: To infer dynamic control information on a controlled object. An inferring device includes one or more memories and one or more processors. The one or more processors are configured to: input at least data about a state of a controlled object and time-series control information for controlling the controlled object, into a network trained by machine learning; acquire predicted data about a future state of the controlled object controlled based on the time-series control information via the network into which the data about the state of the controlled object and the time-series control information have been input; and output new time-series control information for controlling the controlled object to bring the future state of the controlled object into a target state based on the predicted data acquired via the network.

Abstract: A self-position estimation apparatus comprises a camera part; an environmental data storage part that stores information that changes according to the position of a mobile object on a travel route in association with the position information thereof; a first estimation part that estimates the position of mobile object from sensor data containing information that changes according to the position of mobile object and the information in the environmental data storage part; a second estimation part that estimates the self-position of mobile object from a known object included in an image inputted from the camera part; a weighting determination part that determines weighting to the position estimation results by the first estimation part and the second estimation part; and a self-position calculation part that calculates a self-position by linearly combining the self-positions estimated by the first estimation part and the second estimation part using the result from the weighting determination part.

Abstract: An object assessment device determines whether or not an object is present within a predetermined range, and has a processor configured to determine whether or not first object detection information detected at a first detection time point and second object detection information detected at a second detection time point after the first detection time point, are detecting a same object, determine whether or not a predetermined region including a location of the vehicle at the second detection time point satisfies predetermined terrain condition, when it has been determined that the same object has not been detected, determine that the object is present within the predetermined range from the vehicle, when it has been determined that the same object has been detected, or when it has been determined that the predetermined region including the location of the vehicle satisfies the predetermined terrain condition, and give notification of the assessment result.

Abstract: A parking assistance method includes: detecting, from an image of surroundings of a vehicle, a first target object position; detecting a first image capturing situation when the image is captured; with respect to a second target object position of a target object detected from a past image of surroundings of a target parking position in a past, retrieving one or more combinations of a relative positional relationship between the second target object position and the target parking position and a second image capturing situation when the past image is captured; selecting, the relative positional relationship combined with the second image capturing situation having a difference from the first image capturing situation less than or equal to a predetermined difference; and calculating, based on the selected relative positional relationship and the first target object position, a relative position between a current position of the vehicle and the target parking position.

Abstract: Autonomous carriers or totes that include vacuum units are provided. As the totes move or are moved through a warehouse carrying products, they collect debris. The debris can be analyzed at the tote, and actions can be performed based upon the analysis.

Abstract: A vehicle controller includes a processor configured to determine whether to transfer driving control of a vehicle under autonomous driving control to a driver, notify the driver of a transition demand via a notification device in the case where driving control will be transferred to the driver, execute deceleration control of the vehicle at predetermined timing after notification timing of the transition demand, and continue deceleration control of the vehicle even after a predetermined period from the notification timing in the case where the speed of the vehicle decreases to a predetermined speed threshold or less within the predetermined period and where operation by the driver to take over driving is not detected.

Abstract: A system may include a display and a processor communicatively coupled to the display. The processor may be configured to: output, to the display, a view of an airport moving map (AMM); receive aircraft state data and airport surface data; at least based on the current position of the aircraft and at least one factor, obtain commonly used taxi route data from a data structure, the data structure including taxi route data, the taxi route data including information of commonly used taxi routes for the airport, wherein the commonly used taxi route data includes information of a commonly used taxi route for the aircraft on the airport surface; based at least on the commonly used taxi route data, the aircraft state data, and the airport surface data, generate taxi routing guidance content; and output, to the display, the taxi routing guidance content.

Abstract: A hardware processor of a mobile object executes the program stored in a storage device to acquire information indicating a road situation in a traveling direction of a mobile object; to recognize whether the mobile object is moving on a roadway or a predetermined region different from the roadway; to recognize presence of a contact portion between the sidewalk and the predetermined region in the traveling direction of the mobile object; to control the speed of the mobile object at least partially, and limit a speed at which the mobile object is moving on the roadway to a first speed and limit a speed at which the mobile object is moving on a sidewalk to a second speed slower than the first speed; and to bring a speed of the mobile object closer to the second speed when the mobile object is moving on the roadway, the contact portion is recognized within a predetermined range from the mobile object, and the road situation is in a predetermined state.

Abstract: An automatic guided vehicle system includes an automatic guided vehicle and a docking station for docking with the automatic guided vehicle. One of the automatic guided vehicle and the docking station has a pair of first vertical guide surfaces and the other has a pair of rows of rollers. The first vertical guide surfaces extend a predetermined length along a first horizontal direction and are spaced by a predetermined distance in a second horizontal direction perpendicular to the first horizontal direction. The rows of rollers are in rolling contact with the first vertical guide surfaces to guide the automatic guided vehicle to a predetermined position in the docking station.

Abstract: Techniques for item delivery and delivery systems including autonomous delivery vehicles used to identify optimal delivery locations and/or deliver physical items to pedestrians is discussed herein. In some examples, a fleet management system may receive sensor data captured from sensor devices of vehicles of a fleet of vehicles while traversing within an environment. The fleet management system may use the sensor data to identify region(s) of the environment within which a threshold number of pedestrians may be located. Based on identifying a region having a threshold number of pedestrians, the fleet management system may determine how many vehicles of the fleet of vehicles are available to render delivery services to the pedestrians within the region. In some examples, the fleet management system may identify an available vehicle of the fleet of vehicles to which instructions may be sent. Such instructions may include transporting items to pedestrians within the region.