What Is Tele Assistance In Autonomous Driving?
What is Tele Assistance?
Tele Assistance, also known as indirect control, is one part of ‘teleoperation,’ which is simply when a human can monitor, assist and drive an autonomous vehicle (AV) from a remote location. Teleoperation is sometimes interpreted quite narrowly to include only direct methods of control. Ottopia uses a broader definition of teleoperation, encompassing direct and indirect methods of control.
Tele assistance is an indirect control method in teleoperation to assist a car by choosing a path to take, drawing a new path, or overriding a policy because of the situation. Tele assistance is used mainly for vehicles on public roads, such as Robotaxis and shuttles. However, it can also be used on vehicles that drive off public roads, such as in airports, mines, and agricultural areas, although these vehicles are also equipped with remote control capabilities.
An Example Situation Which Requires Tele Assistance
Imagine you catch an autonomous taxi on your way home from work. You relax in the back and check unread messages as the taxi begins to drive through a construction zone.
Outside the vehicle, it’s a busy afternoon as most workers are heading home. Pedestrians and cyclists zip around, not always where they should. The construction site forces some pedestrians and most cyclists to weave around cars and cones. Orange cones, which are mutilated or knocked over, ‘create’ a new lane, contradicting old lane markings from last spring season.
Inside the vehicle, high-end hardware allows the vehicle to sense this entire environment. With LiDAR sensors, radars, cameras, and multiple other sources of information, the vehicle is prepared to respond to almost anything that crosses its path.
At this point, the inherent limitations of teleoperation make it exceedingly difficult, and perhaps even dangerous, for assistance to be provided by a remote operator through direct control. AV technology can help the vehicle drive up to the construction site and away from it, all independently. However, it may not know how to navigate through the complex scenario described above.
If direct control via an operator is not ideal, and there is no one in the car to drive it manually, how can the AV get through this situation?
The Solution: Tele Assistance
Tele Assistance is the answer to this problem. Once the AV recognizes the fact that it needs some assistance, the vehicle begins to slow down and triggers a request for help from a remote operator with some path options. The operator then promptly chooses the correct path for the vehicle to take. The vehicle will adhere to the instructions from the operator, but will ultimately drive by itself should an unknown hazard show up which the operator failed to see.
This is a time where indirect control methods, or tele assistance, are the safest and most effective way for the remote operator to resolve the situation for the autonomous vehicle. To ensure safety, this is how the vehicle operates:
- The movement of the vehicle takes place from within the vehicle itself. Although insights or permissions came from a trained remote operator, all calculations required for navigating the vehicle past the construction site come from the AV algorithm technology.
- This method minimizes any degradation from the system: all sensors and safety measures remain engaged as the vehicle executes any maneuvers.
- Throughout the intervention, the vehicle’s algorithms are allowed to respond to any obstacles, moving objects, or new events occurring in real-time, regardless of the initial input from the remote operator.
Why Is Direct Control Not Preferred Here?
In simple terms, it would be a mistake for this human operator to take direct control of the vehicle and negate the autonomy stack’s cutting-edge visibility and command execution. The AV technology is so clear-cut that it can notice things before a remote operator can. Furthermore, tele assistance is also more time effective as the operator can pick the path, and move onto the next scenario rather than having to be involved in the entire driving process.
Simply drawing the path and providing indirect assistance is enough for the AV to operate alone. Furthermore, the operator might easily handle this situation but there is the added latency from the remote connection, affecting the ability to react and navigate the situation safely.
What Is Exactly Involved In Tele Assistance?
As demonstrated above, an AV that comes across an unusual road situation triggers a request from a teleoperator. Path choosing is the first option that tele assistance offers. Suppose a vehicle comes across an unknown location. In that case, a teleoperator can review the vehicle’s choices, choose the right one, and be assured that the AV technology will guide the car accurately on the correct path.
What if all path options that are given to the teleoperator are not sufficient? In this situation, the teleoperator can draw a new route on their tablet, translating to the AV as the correct path to take. Path drawing is helpful to guide the AV around an obstacle that it may not know how to overcome otherwise.
This is the final part of tele assistance that a teleoperator may implement. There are specific policies that are installed in AV technology, such as ‘don’t run over any object of any kind.’ However, what if there is a cardboard box lying on the road of the chosen path? In this situation, the teleoperator can view, via cameras on the car, that the object is a simple cardboard box; so in this situation a ‘policy override’ button can be clicked, giving the green light for the car to drive over such an object.
This doesn’t mean the policy is suddenly forgotten. The AV understands that a policy override is a one time change to the rules, it isn’t permanent.
These are three of the main methods of tele assistance. However, there are more ways that include go/no go and classification override. Each domain can require its own style of commands and they need to be developed on a case by case basis.
How Does Ottopia Ensure Safety in Tele Assisted Vehicles?
As impressive as this tele assistance sounds, how can you be sure that the network will function properly at the right time? What if the teleoperator doesn’t notice something? What if a last-minute hazard comes up at the last second once the teleoperator has given the new path for the AV to take?
Here are 4 safety principles that we abide by to ensure that tele assistance in teleoperation is as safe as it can be:
Principle 1: Vehicle-Side Calculations
Due to the network limitations that teleoperation encounters in any realistic setting, any safety-critical calculations will not depend on network communications. This is especially important when the connection between vehicle and operator takes place over a public network with competing traffic such as LTE or 5G.
Any final instruction from the system to the vehicle should take place on a computing platform located in the vehicle. The reliability and precision of these calculations are thus independent of network conditions.
Principle 2: Take Full Advantage of System Capabilities
In AV technology, vast resources are dedicated to ensuring that all the system’s key components are optimized for safety and performance in any given domain. It follows that any assistance provided to a vehicle remotely should also take advantage of such capabilities to the greatest extent possible. Doing so can compensate for the differences in situational awareness compared to in-vehicle driving.
Simply put, if the vehicle has smarts (e.g., high-end sensing and computing capabilities), a teleoperation solution must use them.
Principle 3: Algorithm Hierarchy
Should a risk of collision arise during a remote control session, the system is programmed to take control and ignore the operator’s commands. The operator’s commands are, at all times, overseen by the system to ensure maximum safety of the passengers and the vehicle’s surroundings. Therefore, if there is an unforeseen obstacle or last-minute hazard, the algorithm of the vehicle is in control and will drive the car to safety.
Principle 4: Operational Design Domains (ODD) -Specific Safety Protocols
Autonomy comes in many shapes and sizes. Hub-to-hub trucking has very different requirements (and almost certainly, regulatory requirements) than, say, sidewalk delivery robots. The same can be said of autonomous haulage in mines, automated guided vehicles (AGV) in container terminals, or autonomous forklifts in fulfillment centers.
These operating environments — or operational design domains (ODD), to use an industry term — are so diverse that the optimal way of performing safe teleoperation as broadly as possible is to tailor tools and principles as required.
Tele Assistance Is The Best Choice For Most Vehicles
By adhering to Ottopia’s safety principles, indirect methods of control are the most appropriate mechanism for providing remote assistance to an autonomous vehicle in complex environments such as public roads, as they come with higher levels of complexity, unpredictability, and liability.
It is no surprise that for the most complex scenarios our customers and partners prefer integration with their autonomy stack to enable indirect methods of control, combined with a highly reliable and unparalleled network, video, and cyber-security layers — all of which are enabled by Ottopia’s software.