Safe trajectory tracking in uncertain environments

Abstract

The present application discloses safe trajectory tracking in an uncertain environment. The present invention relates to a method for trajectory planning for a vehicle, comprising: obtaining a reference trajectory within a finite time range, the reference trajectory comprising a speed reference over time within the finite time range; determining a backup stop trajectory within the finite time range, the backup stop trajectory having a starting state and terminating at a final state, the final state being defined as a safe state; forming a set of ending states within the finite time range based on at least one predetermined constraint, the set of ending states comprising an ending state corresponding to the starting state of the backup stop trajectory; generating a nominal trajectory within at least a portion of the finite time range based on a constraint control technique, the nominal trajectory depending on the obtained reference trajectory and an ending constraint, the ending constraint defining the nominal trajectory to include an ending state. The constraint control technique comprises a cost minimization control strategy, and the backup stop trajectory from the starting state to the final state is associated with a zero cost.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This patent application claims priority to European Patent Office application serial number 19197329.6, filed on September 13, 2019, entitled “Safe Trajectory Tracking in Uncertain Environments”, assigned to its assignee and expressly incorporated herein by reference. Technical Field

[0003] The present disclosure relates to autonomous driving (AD) and advanced driver assistance systems (ADAS). More specifically, the present disclosure relates to trajectory planning of a vehicle within a drivable area. Background Art

[0004] Today, many vehicles have a variety of driver support features in the form of advanced driver assistance systems (ADAS). Moreover, many of these support features form the basis for current and future automated driving (AD) features. Examples of ADAS features or functions include lane departure warning systems, lane centering, lane keeping assist, driver assistance, lane change assist, parking sensors, pedestrian protection systems, blind spot monitors, adaptive cruise control (ACC), anti-lock braking systems, and the like. These features supplement the traditional driver control of the vehicle with one or more warnings or automatic actions in response to certain situations.

[0005] The development of autonomous vehicles is rapid, and there are often news and demonstrations of impressive technological advances. However, one of the biggest challenges of AD is ensuring that autonomous vehicles can safely plan and execute trajectories.

[0006] More specifically, in modern vehicles, the driver remains a critical component as they are responsible for making many decisions regarding safe operation of the vehicle in terms of speed, steering, obstacle recognition and avoidance, etc. Therefore, in order to realize the vision of fully autonomous operation in the automotive industry, new and improved systems related to various aspects of autonomous driving are required.

[0007] The present disclosure concerns the problem of planning a trajectory for an autonomous vehicle that is stable and, most importantly, safe, not only for the occupants of the vehicle but also for people in the surrounding environment (pedestrians, other vehicles, cyclists, etc.). Summary of the invention

[0008] Therefore, an object of the present disclosure is to provide a method for trajectory planning of a vehicle, a computer-readable storage medium, a control device and a vehicle comprising the control device which alleviate all or at least some of the disadvantages of currently known systems.

[0009] More specifically, the object of the present disclosure is to provide a method for trajectory planning of a vehicle, which allows the vehicle to travel at higher speeds while still ensuring safety in an improved manner compared to currently known solutions. Similarly, the object is to provide a corresponding computer-readable storage medium, a control device and a vehicle comprising the control device.

[0010] This object is achieved by a method for trajectory planning of a vehicle, a computer-readable storage medium, a control device and a vehicle comprising the control device as defined in the appended claims.In the present context, the term "exemplary" is to be understood as serving as an example, instance or illustration.

[0011] According to a first aspect of the present disclosure, a method for trajectory planning of a vehicle is provided. The method includes obtaining a reference trajectory within a finite time range, wherein the reference trajectory includes a speed reference over time within the finite time range. In addition, the method includes determining a spare stop trajectory within the finite time range. The spare stop trajectory has a starting state and terminates at a final state, wherein the final state is defined as a safe state. The method further includes forming a set of ending states within the finite time range based on at least one predetermined constraint, wherein the set of ending states includes an ending state corresponding to the starting state of the spare stop trajectory. In addition, the method includes generating a nominal trajectory within at least a portion of the finite time range based on a constraint control technique, wherein the nominal trajectory depends on the obtained reference trajectory and the ending constraint. The ending constraint defines that the nominal trajectory includes the ending state. The constraint control technique includes a cost minimization control strategy, and the spare stop trajectory from the starting state to the final state is associated with a zero cost. The proposed method provides a trajectory planning solution that can achieve a good compromise between stability (capable of accurately tracking a reference) and safety (the finite time range always contains the spare stop trajectory) based on a constraint control technique (such as, for example, a model predictive control (MPC) framework).

[0012] According to the proposed method, the backup stop component is mainly used to ensure that the vehicle can reach a safe state (e.g., a standstill) within a limited time range. In addition, by forming a set of end states and enforcing the nominal trajectory planning to always include end states from this set, the cost associated with the backup stop trajectory can be mitigated. In the absence of the end state set, the traditional MPC-based control strategy will sacrifice stability (i.e., deviate from the reference trajectory) because of the cost associated with the backup stop trajectory.

[0013] However, the inventors have realised that if an ending constraint is set, i.e. a constraint that the generated nominal trajectory must include one of the states from the ending set of states, then the trajectory can be considered "safe" (for the duration of a finite time range). The advantageous effect is then that the backup stop trajectory need not incur any costs, thus not affecting the nominal part of the generated trajectory, and stability can be achieved. However, if we encounter a situation where we must start executing the backup stop trajectory, then the ending set of states starts to contain only states outside the "reference trajectory", and since the generated trajectory must include one of the states from the ending set, costs are incurred and the nominal trajectory starts to deviate from the reference trajectory (i.e. stability is sacrificed for safety).

[0014] In the present context, a trajectory can be understood as a time-dependent path. Considering the one-dimensional (1D) case where our reference parameter is the velocity parameter of the vehicle, a trajectory is then defined as the velocity value (discrete or continuous) that the vehicle exhibits or will exhibit over a specified period of time.

[0015] The term "limited time range" should be understood as a time range of a limited length, preferably a time range of a predetermined length (e.g., 5 seconds, 10 seconds, 15 seconds). In the case where the length of the limited time range is 10 seconds, the limited time range extends from the current time point (t=0) to the end time (t=t end ), so t end = 10 seconds. The finite time horizon can also be called the prediction horizon.

[0016] According to a second aspect of the disclosure, there is provided a (non-transitory) computer-readable storage medium storing one or more programs, the one or more programs being configured to be executed by one or more processors of a vehicle control system, the one or more programs comprising instructions for performing a method according to any one of the embodiments disclosed herein. For this aspect of the disclosure, there are similar advantages and preferred features as the first aspect of the disclosure discussed previously.

[0017] As used herein, the term "non-transitory" is intended to describe computer-readable storage media (or "memory") that do not include propagating electromagnetic signals, but is not intended to otherwise limit the types of physical computer-readable storage devices encompassed by the term computer-readable media or memory. For example, the term "non-transitory computer-readable medium" or "tangible memory" is intended to encompass types of storage devices that do not necessarily store information permanently, including, for example, random access memory (RAM). Program instructions and data stored in a non-transitory form on a tangible computer-accessible storage medium may be further transmitted via a transmission medium or signals such as electrical, electromagnetic, or digital signals that may be transmitted via a communication medium such as a network and / or wireless link. Thus, as used herein, the term "non-transitory" is a limitation on the medium itself (i.e., tangible, not a signal), not a limitation on the persistence of data storage (e.g., RAM vs. ROM).

[0018] According to a third aspect of the present disclosure, a control device for trajectory planning of a vehicle is provided. The control device includes a control circuit configured to obtain a reference trajectory within a limited time range. The reference trajectory includes a speed reference over time within the limited time range. The control circuit is further configured to determine a spare stop trajectory within the limited time range, wherein the spare stop trajectory has a starting state and a final state. The final state is defined as a safe state. Further, the control circuit is configured to form a set of ending states within the limited time range based on at least one predetermined constraint. The set of ending states includes at least one ending state corresponding to the starting state of the spare stop trajectory. Further, the control circuit is configured to generate a nominal trajectory within at least a portion of the limited time range based on a constraint control technique. The nominal trajectory depends on the obtained reference trajectory and the ending constraint. The ending constraint defines that the nominal trajectory includes an ending state. In addition, the constraint control technique includes a cost minimization control strategy, and wherein the spare stop trajectory from the starting state to the final state is associated with a zero cost. For this aspect of the present disclosure, there are similar advantages and preferred features to the first aspect of the present disclosure discussed previously.

[0019] Other embodiments of the present disclosure are defined in the dependent claims. It should be emphasized that when used in this specification, the term "comprising" is used to specify the presence of stated features, integers, steps or components. It does not exclude the presence or addition of one or more other features, integers, steps, components or groups thereof.

[0020] These and other features and advantages of the present disclosure will be further elucidated below with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Other objects, features and advantages of the embodiments of the present disclosure will become apparent from the following detailed description with reference to the accompanying drawings, in which:

[0022] Figure 1 It is a schematic flowchart of a method for trajectory planning of a vehicle according to an embodiment of the present disclosure.

[0023] Figure 2 Two schematic diagrams illustrating an exemplary embodiment of a method for trajectory planning of a vehicle according to an embodiment of the present disclosure are shown.

[0024] Figure 3 A schematic diagram illustrating an exemplary embodiment of a method for trajectory planning of a vehicle according to an embodiment of the present disclosure is shown.

[0025] Figure 4 is a schematic side view of a vehicle having a control device for lane-level map matching of a vehicle traveling on a road according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] Those skilled in the art will appreciate that the steps, services, and functions described herein may be implemented using separate hardware circuits, using software working in conjunction with a programmed microprocessor or general purpose computer, using one or more application specific integrated circuits (ASICs), and / or using one or more digital signal processors (DSPs). It will also be appreciated that when the present disclosure is described in terms of a method, it may also be embodied in one or more processors and one or more memories coupled to the one or more processors, wherein the one or more memories store one or more programs that, when executed by the one or more processors, perform the steps, services, and functions disclosed herein.

[0027] In the following description of exemplary embodiments, the same reference numerals denote the same or similar components.

[0028] Figure 1 A schematic flow chart representation of a method 100 for controlling an advanced driver assistance feature (ADAS) or automated driving (AD) feature of a vehicle traveling on a road segment according to an example embodiment of the present disclosure is shown. More specifically, the method 100 provides a trajectory tracking / generation feature for an autonomous or semi-autonomous vehicle (i.e., a vehicle equipped with ADAS). In this context, a vehicle will be interpreted broadly and includes cars, buses, trucks, and construction vehicles.

[0029] Even though the present disclosure focuses on vehicles, and in particular, sedans, the teachings herein are applicable to other industries that utilize trajectory planning, such as, for example, robotics, avionics, etc.

[0030] The method 100 includes obtaining 101 a reference trajectory within a limited time range. The reference trajectory includes at least a velocity reference (v ref). Alternatively or additionally, the reference trajectory may include a position reference (p ref ), yaw reference (Θ ref ), acceleration reference (a ref )wait.

[0031] The position reference is preferably in the form of a set of x-positions and y-positions in the road surface. These alternative or additional parameters can, for example, be used to generate steering inputs for following a target path. The reference trajectory can be given based on the current traffic scenario (e.g., based on the vehicle's geographic location and the surrounding traffic) and can be predefined or can be dynamic based on sensor observations. Thus, the reference trajectory can be part of (local) "map data" and can be obtained from an external entity (which is part of a vehicle management system) for example via an external network (e.g., a cellular network).

[0032] The term limited time range is to be understood as a time range of a limited length, preferably a time range of a predetermined length (e.g., 5 seconds, 10 seconds, 15 seconds). In the case where the length of the limited time range is 10 seconds, the limited time range extends from the current time point (t=0) to the end time (t=t end ), so t end = 10 seconds. The finite time horizon can also be called the prediction horizon.

[0033] Further, the method 100 includes determining 102 one or more alternate stop trajectories (also referred to as safe stop trajectories) within a limited time range. The alternate stop trajectory has a starting state and terminates in a final state. The final state of each alternate stop trajectory is defined as a safe state. In an example embodiment of the present disclosure, the final state is a state where the speed of the vehicle is zero. However, the method can also be used in other applications (e.g., robotics), in which case the safe state can be, for example, when the robot unit (e.g., a robot arm) reaches a predetermined position and / or configuration (e.g., a fully opened handle). According to the background of one embodiment, the alternate stop trajectory is in the maximum deceleration trajectory of the vehicle, that is, transitioning from a first vehicle speed to a standstill as quickly as possible. Therefore, the alternate stop trajectory can be interpreted as an emergency stop action, in which the priority is to stop the vehicle as quickly as possible while reducing considerations for occupant comfort.

[0034] Further, the method 100 comprises forming 103 a terminal set of states within a limited time range based on at least one predetermined constraint. The terminal set of states comprises at least one terminal state corresponding to a start state of the backup stop trajectory. The at least one predetermined constraint may be, for example, a road geometry of the vehicle's surroundings, a maximum deceleration capability of the vehicle, or weather data.

[0035] In other words, vehicle dynamics, road conditions, and / or reference trajectory speed may be considered when forming the end state set. In more detail, each state in the end state set must satisfy the conditions that enable the final state (safe state, such as zero vehicle speed) to be reached. Therefore, in order to ensure that this is possible, various vehicle characteristics may be considered, such as vehicle weight, braking capability, road conditions, etc. These various characteristics define one or more constraints used to form the end state set.

[0036] Next, a nominal trajectory within at least a portion of the limited time horizon is generated 104 based on a constrained control technique. The constrained control technique may be based on a model predictive control (MPC) framework. However, other techniques, such as reference and command governors, may also be employed. The nominal trajectory depends on the obtained reference trajectory and an end constraint, wherein the end constraint defines that the control trajectory includes states from an end state set. Furthermore, the constrained control technique includes a cost minimization control strategy, and wherein an alternate stop trajectory from the end state to the final state is associated with a zero cost.

[0037] In other words, the proposed method 100 provides a trajectory planning solution that ensures that there is always a "safe trajectory" (i.e., an alternate stop trajectory) within a finite time horizon without sacrificing stability. In more detail, as long as we do not need to "follow" the alternate stop trajectory, the controller is "settling down" (i.e., the nominal trajectory converges to the reference trajectory) because the alternate stop trajectory is not associated with any cost. We understand the term "not needing to follow the alternate stop trajectory" to mean that there is no need to initiate the alternate stop trajectory at a subsequent time sample (which will be further described in Figure 2 for examples and explanations).

[0038] An advantage of the proposed approach is then that a trajectory planning module can be implemented that achieves a good compromise between stability (being able to accurately track the reference) and safety (the limited time horizon always includes an executable alternative stop trajectory, which is guaranteed by including the end state in the generated target trajectory (nominal trajectory)). In other solutions, the alternative stop trajectory is included in the limited time horizon in order to be able to ensure safety. However, including the "alternative stop trajectory" only in the limited time horizon will force the trajectory planning module to deviate from the reference trajectory in these existing solutions. This is because emergency braking actions are associated with high costs, and since the cost minimization control strategy finds the optimal trajectory by minimizing the cost, it will sacrifice convergence to the reference trajectory to avoid being exposed to an "emergency braking situation". Even if the alternative stop trajectory is not the maximum deceleration trajectory (emergency braking action), it still incurs costs and weakens the ability to stabilize against the high-speed reference trajectory.

[0039] Furthermore, guaranteeing safety for an infinite time range may be impossible, so the challenge contemplated in the present disclosure is how to "guarantee" safety for a limited time range without sacrificing stability (e.g., driving unreasonably slowly).

[0040] The inventors have realized that the probability of being exposed to an "emergency braking" situation is relatively low, so we do not need a trajectory planning system that generates a target trajectory based on the assumption that an emergency stop exists in each planning horizon. Therefore, by forming a set of end states and ensuring that the nominal trajectory always includes one of these end states, but without the cost of allocating an alternate stop trajectory, the trajectory planning module is allowed to focus on stabilizing the nominal trajectory relative to the reference trajectory, and the trajectory planning module is still able to ensure that a "safe state" can be reached within a limited time horizon.

[0041] In other words, the method generates a nominal trajectory based on a constrained control policy that only needs to include one of the end states in order to be able to claim (with high probability) that the generated trajectory is "safe" within the prediction horizon (limited time horizon). Therefore, the backup stop trajectory does not need to include any cost, since the definition of the end state is that as long as the vehicle can reach the end state, it can also execute the backup stop trajectory (thus reaching a safe state). Therefore, it can be said that the set of end states provides the necessary degrees of freedom to stabilize the nominal trajectory without sacrificing the safety of the trajectory planning module.

[0042] Executable instructions for performing these functions are optionally included in a non-transitory computer-readable storage medium or other computer program product configured to be executed by one or more processors.

[0043] Figure 2 Two schematic diagrams illustrating an exemplary embodiment of a method for trajectory planning of a vehicle traveling on a road segment are shown. In more detail, the top diagram shows the generated trajectory at time t=k, while the bottom diagram shows the generated trajectory at the subsequent time step t=k+3, i.e. three 25 sampling times later. The diagram includes a speed reference v indicated by a dashed line 21 ref The sampling time is set according to the application and specifications and can be, for example, 1 ms, 10 ms or 100 ms, corresponding to a sampling rate of 1000 Hz, 100 Hz or 10 Hz, respectively.

[0044] Therefore, once the reference trajectory 21 is obtained within a limited time range, at least one alternative stop trajectory within a limited time range is determined. The alternative stop trajectory has a starting state 22 and ends in a final state 23, which is defined as a safe state (here, a state where the speed of the vehicle is zero). In this context, a vehicle state in which the vehicle is not moving is considered "safe". This is because a stationary vehicle is not responsible for being hit by other road users. We can compare this with other technical fields, for example, a closed chemical reactor is generally considered safe, a closed electrical circuit can be considered safe or a ship docked in a port can be considered safe.

[0045] Furthermore, a set of end states 22 is formed within a finite time range based on at least one constraint. The set of end states includes at least one end state 22 corresponding to the starting state of the alternative stop trajectory. It can be said that the end state is formed based on a set of constraints or criteria (e.g., vehicle dynamics, road geometry, passenger comfort, etc.) to ensure that a safe state can be reached. Thus, one or more end states may be one or more vehicle states (position, velocity, and azimuth) that the vehicle must assume if the vehicle is to reach a safe state according to the estimated trajectory. Alternative alternative stop trajectories are represented by dashed lines between the corresponding end states 22 and the final state 23. In the illustrated example, the selected end state corresponds to the starting state of the alternative stop trajectory in the form of a maximum deceleration trajectory.

[0046] In more detail, the outer boundary of the set of end states is the start state of the maximum deceleration trajectory. Typically, this trajectory allows the highest speed of the start state. In an exemplary embodiment, the backup stop trajectory is a maximum deceleration trajectory (which may also be referred to as an emergency brake trajectory). For the backup stop trajectory in the form of a maximum deceleration trajectory, little or no consideration is given to passenger comfort. By selecting the backup stop trajectory in the form of a maximum deceleration trajectory, the nominal trajectory is most likely to be stable at a high reference speed.

[0047] In addition, a nominal trajectory (here, represented by a vehicle speed parameter) is generated within at least a portion of a limited time range based on the constraint control technique. The nominal trajectory depends at least on the obtained reference trajectory 21 and the end constraint. The end constraint defines that the nominal trajectory includes a state 22 from a set of end states 24. As described above, the constraint control technique includes a cost minimization control strategy, and wherein an alternative stop trajectory from a starting state (corresponding to the end state) to a final state is associated with a zero cost. Naturally, depending on the intended application and the specifications associated therewith, the nominal trajectory can be further based on other constraints. For example, various vehicle parameters can define additional constraints (maximum acceleration, turning radius, vehicle size, etc.), and in addition, user preferences can also indicate other constraints related to comfort (e.g., lateral acceleration, vertical acceleration, etc.).

[0048] Furthermore, the limited time range includes a nominal time range (defined as a first portion of the limited time range) and an alternate stop time range (defined as a second portion of the limited time range). Preferably, the nominal trajectory is included in the nominal time range and the alternate stop trajectory is included in the alternate stop time range.

[0049] The ratio between the nominal time range and the alternate stopping range can be set based on user preferences and / or based on the current surroundings of the vehicle (e.g., highway cruising, rush hour traffic in a densely populated area, etc.). For example, if the limited time range is ten seconds (or any other suitable value depending on the performance of the vehicle's perception system), the nominal time range can be five seconds and the alternate stopping time range can be five seconds, i.e., a 50 / 50 ratio. However, other ratios are also possible, such as 60 / 40, 40 / 60, 70 / 30, 30 / 70, etc.

[0050] As mentioned before, improvements in occupant safety can be achieved as long as there are alternate stop trajectories considered in the finite time horizon. Moreover, as long as we do not need to execute the alternate stop trajectory, the controller (operating under the constraint control technique) is "stable" (e.g., allowing the vehicle to converge to a reference speed). Therefore, during "normal" operation, during each sample / update / iteration of trajectory planning, the safe state 23 is set at the end of the finite time horizon. This is further illustrated in the bottom figure, where at each update, the finite time horizon is simply moved forward in time. An important aspect to consider is that we have "planned" an alternate stop trajectory within the finite time horizon, i.e., the vehicle can reach a safe state within the finite time horizon. Therefore, through the proposed method, it can be considered that it is ensured that the vehicle can always reach a "safe state" within the planning horizon and still maintain the good stability performance of the nominal trajectory planning.

[0051] However, if the vehicle sensors detect an unexpected obstacle that blocks the vehicle's target path (e.g., a pedestrian crossing the street), the safe state, and therefore the alternate stop trajectory, will propagate "inward" into the nominal time horizon, thereby beginning to incur costs in generating the nominal trajectory. More specifically, the formation (or shape) of the set of ending states will be affected, which in turn will affect the nominal trajectory. In that case, the nominal trajectory will begin to deviate from the reference trajectory 21 (e.g., the vehicle speed will be reduced) to account for the costs associated with the new set of ending states 22. This is in the Figure 3 It is shown in Figure 3 A diagram illustrating a method for trajectory planning of a vehicle is shown.

[0052] More specifically, Figure 3 The figure shows that in Figure 2The bottom diagram of FIG. 1 continues with an example scenario or situation in which an obstacle is detected (e.g., by a perception system of the vehicle) in a subsequent time sample 25. Detection of the obstacle forces the trajectory planning module to plan a trajectory that is longer than the previous time sample 25 ( Figure 2 The vehicle reaches the standstill state 23 faster than the reference speed (v ref ) reaches a safe state with maximum deceleration, and the end state 22 closest to the reference trajectory 21 differs from the reference trajectory 21 sufficiently to cause the trajectory planning module to sacrifice stability for safety when generating the nominal trajectory (indicated by the double arrow 26).

[0053] In other words, the end state set 24 includes only states outside the "reference trajectory", and because the generated nominal trajectory must include a state from the end state set 24, a cost is incurred and the nominal trajectory begins to deviate from the reference trajectory 21 (i.e., stability is sacrificed for safety).

[0054] Furthermore, in another exemplary embodiment, the method includes obtaining a predicted trajectory of at least one dynamic object (e.g., a pedestrian, a cyclist, another vehicle, etc.) located in the surroundings of the vehicle. For example, the predicted trajectory can be formed based on sensor data obtained from at least one sensor and a predetermined (motion) model. Then, a set of end states is further formed based on each predicted trajectory. The sensor data can originate from onboard sensors of the ego vehicle (e.g., radar, LIDAR, camera, etc.), connected infrastructure equipment (e.g., traffic lights, traffic cameras, toll systems, etc.), and / or other vehicles accessible via a vehicle-to-vehicle (V2V) solution. Therefore, at each sampling time, the current traffic conditions are taken into account when filling the end state set with the end state. Figure 2 and Figure 3 The illustrated example of shows a case where the nominal trajectory includes an end state 22 corresponding to an alternate stop trajectory in the form of maximum deceleration, and those skilled in the art will readily appreciate that other end states 22 within the end state set may be selected for inclusion in the nominal trajectory. The selection of which end state to include in the nominal trajectory generation may be based on predetermined preferences and specific scenarios (user settings, environmental data, etc.). In addition, as Figure 2 and Figure 3 As shown, the "shape" of the end state set 24 is dynamic, and it can be said that the "shape" of the end state set 24 affects the "stability" of the nominal trajectory (i.e., the ability to accurately track the reference trajectory 21). Therefore, the shape of the end state set may be affected by external objects that appear or disappear in the vehicle's surroundings.

[0055] Figure 41 is a schematic side view of a vehicle 1 including a control device 10 for controlling an advanced driver assistance (ADAS) feature or an autonomous driving (AD) feature of the vehicle 1 traveling on a road segment. The ADAS or AD feature may be, for example, a trajectory planning feature. The vehicle 1 further includes a perception system 6, an inertial measurement unit (IMU) 7, and a positioning system 5. In the current context, the perception system 6 should be understood as a system responsible for acquiring raw sensor data from sensors 6a, 6b, 6c such as cameras, LIDAR and RADAR, ultrasonic sensors, and converting the raw data into scene understanding. The positioning system 5 is configured to monitor the geographic location and direction of the vehicle and may take the form of a global navigation satellite system (GNSS) such as GPS. However, alternatively, the positioning system may be implemented as a real-time kinematic (RTK) GPS to improve accuracy. The IMU 7 should be understood as an electronic device configured to measure the inertial motion of the vehicle 1. The IMU 7 typically has six degrees of freedom, three accelerometers and three gyroscopes.

[0056] The control device 10 includes one or more processors 11, a memory 12, a sensor interface 13, and a communication interface 14. The processor 11 may also be referred to as a control circuit 11 or a control circuit system 11. The control circuit 11 is configured to execute instructions stored in the memory 12 to perform a method for trajectory planning for a vehicle according to any one of the embodiments disclosed herein. In other words, the memory 12 of the control device 10 may include one or more (non-temporary) computer-readable storage media for storing computer-executable instructions, which, when executed by one or more computer processors 11, may, for example, cause the computer processor 11 to perform the techniques described herein. Optionally, the memory 12 includes a high-speed random access memory, such as DRAM, SRAM, DDR RAM, or other random access solid-state storage devices; and, optionally, includes a non-volatile memory, such as one or more disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices.

[0057] In more detail, the control circuit 11 is configured to obtain a reference trajectory within a limited time range, the reference trajectory including a speed reference over time within the limited time range. The reference trajectory can be obtained from a remote server 2, for example, from a local memory unit 12 or an external network accessible via an onboard antenna 8. In addition, the control circuit 11 is configured to determine a backup stop trajectory within a limited time range. The backup stop trajectory has a starting state and a final state, the final state being defined herein as a safe state.

[0058] Next, the control circuit 11 is configured to form a set of end states within a limited time range based on at least one predetermined constraint. The set of end states includes at least one end state corresponding to the starting state of the backup stop trajectory. In addition, the control circuit 11 is configured to generate a nominal trajectory within at least a portion of the limited time range based on a constraint control technique (e.g., MPC). The nominal trajectory depends on the obtained reference trajectory and the end constraint. The end constraint defines that the nominal trajectory includes the end state. The constraint control technique includes a cost minimization control strategy, and the backup stop trajectory from the starting state to the final state is associated with a zero cost.

[0059] Furthermore, the vehicle 1 can be connected to an external network 2 (e.g. for map data retrieval) via, for example, a wireless link. The same or some other wireless links can be used for communication with other vehicles in the vicinity of the vehicle or with local infrastructure elements. Cellular communication technology can be used, for example, for long-range communication to external networks, and if the cellular communication technology used has low latency, it can also be used for communication between vehicles, vehicle-to-vehicle (V2V) and / or vehicle-to-infrastructure V2X. Examples of cellular radio technologies are GSM, GPRS, EDGE, LTE, 5G, 5G NR, etc., also including future cellular solutions. However, in some solutions, short and medium-range communication technologies are used, for example, wireless local area networks (LANs) such as solutions based on IEEE 802.11. ETSI is developing cellular standards for vehicle communications, and 5G is considered a suitable solution, for example, due to low latency and efficient handling of high bandwidths and communication channels.

[0060] The present disclosure has been presented above with reference to specific embodiments. However, other embodiments in addition to the above-described embodiments are possible and within the scope of the present disclosure. Within the scope of the present disclosure, method steps different from the above-described method steps may be provided for performing the method by hardware or software. Therefore, according to an exemplary embodiment, a non-transitory computer-readable storage medium is provided, which stores one or more programs configured to be executed by one or more processors of a vehicle control system, and the one or more programs include instructions for performing a method according to any one of the above-described embodiments. Alternatively, according to another exemplary embodiment, a cloud computing system may be configured to perform any of the methods presented herein. The cloud computing system may include distributed cloud computing resources that jointly perform the methods presented herein under the control of one or more computer program products.

[0061] In general, computer-accessible media may include any tangible or non-transitory storage media or memory media, such as electronic, magnetic, or optical media, for example, a disk or CD / DVD-ROM coupled to a computer system via a bus. As used herein, the terms "tangible" and "non-transitory" are intended to describe computer-readable storage media (or "memory") that do not include propagating electromagnetic signals, but are not intended to otherwise limit the types of physical computer-readable storage devices encompassed by the term computer-readable media or memory. For example, the term "non-transitory computer-readable media" or "tangible memory" is intended to cover types of storage devices that do not necessarily store information permanently, including, for example, random access memory (RAM). Program instructions and data stored in a non-transitory form on a tangible computer-accessible storage medium may be further transmitted via a transmission medium or signals, such as electrical, electromagnetic, or digital signals, that may be transmitted via a communication medium such as a network and / or a wireless link.

[0062] Processor 11 (associated with control device 10) can be or include any number of hardware components for performing data or signal processing or for executing computer code stored in memory 12. Device 10 has associated memory 12, and memory 12 can be one or more devices for storing data and / or computer code for completing or facilitating the various methods described in this specification. The memory may include volatile memory or non-volatile memory. Memory 12 may include a database component, an object code component, a script component, or any other type of information structure for supporting the various activities of this specification. According to an exemplary embodiment, any distributed or local storage device can be used with the system and method of this specification. According to an exemplary embodiment, memory 12 (e.g., via a circuit or any other wired, wireless or network connection) is communicatively connected to processor 11 and includes computer code for performing one or more processes described herein.

[0063] It will be appreciated that the sensor interface 13 may also provide the possibility of acquiring sensor data directly or via a dedicated sensor control circuit system 6 in the vehicle. The communication / antenna interface 14 may further provide the possibility of sending the output to a remote location (e.g., a remote operator or a control center) by means of an antenna 5. In addition, some sensors in the vehicle may communicate with the control device 10 using a local network setup such as a CAN bus, I2C, Ethernet, optical fiber, etc. The communication interface 14 may be arranged to communicate with other control functions of the vehicle and may therefore also be considered a control interface; however, a separate control interface (not shown) may be provided. Local communications within the vehicle may also be of a wireless type with protocols such as WiFi, LoRa, Zigbee, Bluetooth or similar medium / short range technologies.

[0064] It will therefore be appreciated that parts of the described solution may be implemented in the vehicle, in a system disposed externally to the vehicle, or in a combination of internal and external to the vehicle; for example, in a server in communication with the vehicle, a so-called cloud solution. For example, sensor data may be sent to an external system, and the system performs the steps of determining one or more alternate stop trajectories and forming a set of end states. Different features and steps of the embodiments may be combined in other combinations than those described.

[0065] It should be noted that the word "comprising" does not exclude the presence of other elements or steps than those listed, and the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. It should also be noted that any reference numerals do not limit the scope of the claims, the present disclosure may be implemented at least in part by both hardware and software, and a plurality of "means" or "units" may be represented by the same item of hardware.

[0066] Although the accompanying drawings can show a specific order of method steps, the order of steps may be different from the order depicted. In addition, two or more steps may be performed simultaneously or partially concurrently. This variation will depend on the selected software and hardware system and the designer's choice. All these variations are within the scope of the present disclosure. Similarly, software implementations may be accomplished using standard programming techniques with rule-based logic and other logic to accomplish various connection steps, processing steps, comparison steps, and decision steps. The embodiments mentioned and described above are given as examples only and should not limit the present disclosure. Other solutions, uses, objectives, and functions within the scope of the present disclosure claimed in the patent embodiments described above should be apparent to those skilled in the art.

Claims

1. A method for trajectory planning of a vehicle, the method comprising: Obtaining a reference trajectory within a limited time range, the reference trajectory comprising a speed reference over time within the limited time range; determining an alternate stopping trajectory within the limited time range, the alternate stopping trajectory having a starting state and terminating in a final state, the final state being defined as a safe state; forming an end state set within the limited time range based on at least one predetermined constraint, wherein the end state set includes an end state corresponding to the start state of the backup stop trajectory; generating a nominal trajectory within at least a portion of the limited time range based on a constraint control technique, the nominal trajectory being dependent on the obtained reference trajectory and an end constraint, wherein the end constraint defines that the nominal trajectory includes the end state; and Wherein the constrained control technique comprises a cost minimization control strategy, and wherein the alternate stopping trajectory from the starting state to the final state is associated with a cost of zero.

2. The method according to claim 1, wherein: The backup stop trajectory is associated with a zero cost of the generated nominal trajectory as long as the corresponding end state of the set of end states is included in the nominal trajectory.

3. The method according to claim 1, wherein: The nominal trajectory ends with an ending state from the set of ending states.

4. The method according to claim 1, wherein: The constraint control technique is based on the model predictive control (MPC) framework.

5. The method according to claim 1, wherein: the limited time range includes a nominal time range defined as a first portion of the limited time range and an alternate stop time range defined as a second portion of the limited time range, the first portion preceding the second portion; and Wherein, the nominal trajectory is included in the nominal time range, and the backup stop trajectory is included in the backup stop time range.

6. The method according to claim 5, wherein: The first portion is a first predetermined time period of the limited time range, and wherein the second portion is a second predetermined time period of the limited time range, the first predetermined time period and the second predetermined time period being equal to a total time of the limited time range.

7. The method according to claim 1, wherein: The final state is a state where the speed of the vehicle is zero.

8. The method according to claim 1, wherein: The at least one predetermined constraint includes at least one of road geometry of the vehicle's surroundings, a maximum deceleration capability of the vehicle, and weather data.

9. The method according to claim 1, further comprising: For a first time sample, sending a signal to control a speed of the vehicle based on the generated nominal trajectory; and shifting the limited time range forward in time by one time sample; and The method according to any of the preceding claims is repeated, wherein the finite time range is replaced by a shifted finite time range.

10. The method according to claim 1, further comprising: obtaining a predicted trajectory of at least one dynamic object located in a surrounding environment of the vehicle, wherein the predicted trajectory is based on sensor data obtained by at least one sensor and a predetermined model; and Wherein, the end state set is further formed based on each predicted trajectory.

11. A computer readable storage medium storing one or more programs configured to be executed by one or more processors of a vehicle control system, the one or more programs comprising instructions for executing a method for trajectory planning for a vehicle, the method comprising: Obtaining a reference trajectory within a limited time range, the reference trajectory comprising a speed reference over time within the limited time range; determining an alternate stopping trajectory within the limited time range, the alternate stopping trajectory having a starting state and terminating in a final state, the final state being defined as a safe state; forming an end state set within the limited time range based on at least one predetermined constraint, wherein the end state set includes an end state corresponding to the start state of the backup stop trajectory; generating a nominal trajectory within at least a portion of the limited time range based on a constraint control technique, the nominal trajectory being dependent on the obtained reference trajectory and an end constraint, wherein the end constraint defines that the nominal trajectory includes the end state; and Wherein the constrained control technique comprises a cost minimization control strategy, and wherein the alternate stopping trajectory from the starting state to the final state is associated with a cost of zero.

12. A control device for trajectory planning of a vehicle, the control device comprising a control circuit, the control circuit being configured to: Obtaining a reference trajectory within a limited time range, the reference trajectory comprising a speed reference over time within the limited time range; determining a backup stop trajectory within the limited time range, the backup stop trajectory having a starting state and a final state, the final state being defined as a safe state; forming an end state set within the limited time range based on at least one predetermined constraint, wherein the end state set includes an end state corresponding to the start state of the backup stop trajectory; generating a nominal trajectory within at least a portion of the limited time range based on a constraint control technique, the nominal trajectory being dependent on the obtained reference trajectory and an end constraint, wherein the end constraint defines that the nominal trajectory includes the end state; and Wherein the constrained control technique comprises a cost minimization control strategy, and wherein the alternate stopping trajectory from the starting state to the final state is associated with a cost of zero.

13. A vehicle comprising the control device according to claim 12.

Images